Doctoral Thesis
Refine
Year of publication
- 2020 (13) (remove)
Document Type
- Doctoral Thesis (13) (remove)
Language
- English (13) (remove)
Has Fulltext
- yes (13)
Is part of the Bibliography
- no (13)
Keywords
- BMI (1)
- CEBPD (1)
- Collagen-based biomaterial (1)
- Cystic fibrosis (1)
- FEV1 (1)
- HDAC and BET inhibitor (1)
- Intravenous antibiotic therapy (1)
- Lung function (1)
- Macrophages (1)
- Platelet-rich fibrin (1)
Institute
- Medizin (13) (remove)
Background: Previous studies have demonstrated that CF (Cystic Fibrosis) prognosis is dependent of three major parameters: FEV1 (Forced Expiratory Pressure in one second), BMI (Body Mass Index) and need of intravenous antibiotic therapy. The CF centres of Frankfurt, Germany, and Moscow, Russia, care for cystic fibrosis patients. We decided to investigate and compare both centers from 1990 to 2015. No comparable study has been published so far.
Method: German patient data was collected from the national cystic fibrosis database “Muko.web”. Missing values were extracted from the Hospital Information System. Russian patient data were taken directly from the medical records in Moscow. In a descriptive statistical analysis with Bias and R Studio the values were compared.
Result: A total of 428 patients from Moscow (217 male, 211 female; 348 (81,3%) were P. aeruginosa positive) and 159 patients from Frankfurt (92 male, 67 female; 137 (86,2%) with P. aeruginosa positive) were compared with regard to P. aeruginosa positivity, BMI, FEV1 and need of intravenous antibiotic therapy. CF patients in Moscow stratified by age groups had lower BMI than CF patients in Frankfurt (age 16-18: p=0,003; age 19-22: p=0,004; age 23-29: p<0,001; age 30-35: p<0,001; age 36-66: p=0,024). In a matching pairs analysis including 100 patients from Frankfurt and 100 patients from Moscow for the year 2015 FEV1 was significantly lower in Moscow patients (p<0,001).
Conclusion: BMI, FEV1 and need of intravenous therapy have significant impact on survival and on quality of life of CF patients. A lower BMI and a lower FEV1 result in a worse survival and determine the prognosis. This study showed a significant difference in prognostic parameters between Frankfurt and Moscow in the crosssectional analysis for the year 2015. A further study should evaluate this difference to show whether this difference will be found over a longer period of time.
Endothelial dysfunction plays an important role in different pathological conditions, but whether endothelial cell death contributes to the development and progression of certain pathological conditions is rather unclear. Here we found that endothelial cells undergo cell death during pathologies such as LPS-induced sepsis and in models of hindlimb, renal and cardiac ischemia-reperfusion injury. Analyses of mice lacking endothelial key cell death regulators such as TAK1, RIPK3 and Caspase 8 gave us insight in the role of endothelial cell death in these pathological models. For example, increased endothelial necroptosis along with basal inflammation in lungs of TAK1ECKO mice affects susceptibility to LPS-induced sepsis and mortality, which correlated with elevated IFN-gamma and MIP-2 serum levels. Furthermore, we found that inhibition of RIPK3-mediated endothelial necroptosis could reduce the susceptibility of TAK1ECKO mice to LPS-induced sepsis and mortality. In ischemia or ischemia-reperfusion models, inhibition of RIPK3-mediated endothelial necroptosis did not reduce injury in the heart after ischemia, nor did it have any effect on organ function post-injury in the kidney or the heart. Inhibition of necroptosis also did not alter vascularization processes in hindlimb post-ischemia. Taken together, endothelial necroptosis contributes to increased sepsis severity and progression whereas inhibition of endothelial necroptosis can ameliorate susceptibility to sepsis in the absence of endothelial TAK1. Inhibition of endothelial necroptosis however does not play an important role during ischemia or ischemia-reperfusion induced organ injury.
Bipolar disorder (BD) and major depressive disorder (MDD) are severe mood disorders that belong to the most debilitating diseases worldwide. Differentiating both mood disorders often poses a major clinical challenge, leading to frequent misdiagnoses. Objective biomarkers able to differentiate individuals with BD and MDD therefore represent a psychiatric research field of utmost importance. Recent studies have applied resting-state fMRI paradigms and found promising results differentiating both disorders based on the acquired data. However, most of these studies have focused their efforts on acutely depressed patients. Thus, it remains unclear whether the aberrations remain in a symptomless disease state.
The here presented study addresses these issues by evaluating the ability to differentiate both disorders from one another by conducting a between-group comparison of functional brain network connectivity (FNC) obtained from resting-state fMRI data. Data were collected from 20 BD, 15 MDD patients and 30 age- and gender-matched healthy controls (HC). Graph theoretical analyses were applied to detect differences in functional network organization between the groups on a global and regional network level.
Network analysis detected frontal, temporal and subcortical nodes in emotion regulation areas such as the limbic system and associated regions exhibiting significant differences in network integration and segregation in BD compared to MDD patients and HC. Participants with MDD and HC only differed in frontal and insular network centrality.
These results indicate that a significantly altered brain network topology in the limbic system might be a trait marker specific to BD. Brain network analysis in these regions may therefore be used to differentiate euthymic BD not only from HC but also from patients with MDD.
Current research on medical biomaterials have shown that the physical and chemical characteristics of biomaterials determine the body inflammatory cellular reaction after their implantation. The aim of this study was to evaluate the individual effects of the physical characteristics over the initial biomaterial-cellular interaction and the inflammatory cellular reaction. For this purpose, an equine-derived collagen hemostatic sponge (E-CHS) was modified by pressing and evaluated using ex vivo, in vitro and in vivo methods.
The E-CHS was pressed by applying constant pressure (6.47± 0.85 N) for 2 min using a sterile stainless-steel cylinder and cut in segments of 1cm2. Subsequently, E-CHS and the pressed equine-derived collagen hemostatic sponge (P-E-CHS) were studied as two independent biomaterials and compared to a control group (CG).
A blood concentrate containing inflammatory cells known as platelet rich fibrin (PRF) was used to mimic the initial biomaterial-cell interaction and to measure the absorption coefficient of the biomaterials to liquid PRF (iPAC). Additionally, the biomaterials were cultivated together with PRF for 3 and 6 days to measure the induction of pro-inflammatory cytokines (TNF-α and IL-8). The results were obtained through enzyme-linked immunosorbent assay (ELISA) and histological methods. PRF cultivated without biomaterials served as the CG. Additionally, the biomaterials were evaluated in vivo using a subcutaneous model in Wistar rats and compared to sham operated animals (CG) representing physiologic wound healing. After 3, 15 and 30 days, the explanted samples were evaluated using histochemical and immunohistochemical (IHC) staining using the following markers: CD68 (pan macrophages), CCR7 (pro-inflammatory macrophages, M1), CD206 (pro-wound healing macrophages, M2) and α-Smooth Muscle Actin (α-SMA; vessel identification).
After the mixture of liquid PRF with both biomaterials for 15 minutes, the ex vivo results showed that E-CHS was penetrated by cells, whereas P-E-CHS was cell-occlusive. Additionally, P-E-CHS induced a higher release of pro-inflammatory cytokines compared to liquid PRF alone (CG) and E-CHS after 3 days (P< 0.05). Although the biomaterial was pressed, the difference of the iPAC value did not show statistical differences. In vivo, the CG induced at day 3 a higher inflammatory response compared to the experimental groups (EG) (P< 0.05). The intergroup comparison showed that P-E-CHS induced a higher presence of macrophages (CD68+/CC7+) compared to E-CHS at day 3 (P< 0.05). Only CD68+/CCR7+ mononuclear cells (MNCs) were observed without multinucleated giant cells (MNGCs). After 15 days, the presence of macrophages (CD68+ P<0.01 /CCR7+ P<0.001 /CD206+ P<0.05) reduced considerably in the CG. On the contrary, the inflammatory response increased in the EGs (CD68+/CCR7+). The intergroup comparison showed that this increment was statistically significant when comparing E-CHS and P-E-CHS to the CG at day 15 (P<0.01 and P< 0.05 respectively). At this time point, a reduced number of MNGCs were observed in the EGs. In the CG no MNGCs were observed. Furthermore, E-CHS showed a faster degradation rate and was fully invaded by cells and vessels formed in its interior region. On the other hand, P-E-CHS remained occlusive to cell penetration and vessels were formed only in the periphery. After 30 days, the cellular reaction shifted to a higher number of M2 macrophages (CD260+) in all groups and a reduced presence of CD68+ and CCR7+ MNCs. Both biomaterials degraded and only small fragments were found in the implantation bed surrounded by MNGCs (CCR7+).
These results are of high clinical relevance and show that changes in biomaterial properties have a significant impact on their interaction with the body. They also serve as insight into the possibility to develop versatile biomaterials with different applications. For example, E-CHs can be applied to support hemostasis in a bleeding alveolar socket and P-E-CHs by being cell occlusive and having a delayed degradation rate can be applied for guided bone and tissue regeneration.
BH3 mimetics are novel anticancer therapeutics that induce apoptosis by targeting anti‐apoptotic BCL‐2 proteins. Highly specific inhibitors of the main anti-apoptotic proteins BCL-2, BCL‐XL and MCL‐1 promise new opportunities for the treatment of AML. However, it is currently unclear which of these anti-apoptotic BCL-2 proteins represents the most promising target in AML. Therefore, we investigated the effect of BH3 mimetics targeting either BCL-2 (ABT-199, S55746), BCL-XL (A-1331852) or MCL-1 (S63845) on eleven AML cell lines. Drug sensitivity screening revealed heterogeneous sensitivity towards the different BH3 mimetics, with the best responses observed upon targeting of MCL-1. Selected cell lines that displayed sensitivity towards the specific BH3 mimetics underwent intrinsic apoptosis, which was characterized by loss of mitochondrial membrane potential, exposure of phosphatidylserine and activation of caspases. Furthermore, S63845 turned out to displace BIMS and NOXA from MCL-1 to induce apoptotic cell death. Importantly, the translational relevance of this study was demonstrated by experiments in primary AML blasts, which displayed similar sensitivity towards BH3 mimetics as the cell lines did. Additionally, experiments with nonmalignant cells could confirm the clinical relevance of the MCL-1 inhibitor. There we could show, that S63845 does not cause cytotoxicity on HPCs at efficacious doses.
In conclusion, our findings reveal that the inhibition of BCL-2 proteins, especially MCL-1, by BH3 mimetics can be a promising strategy in AML treatment.
Although immune checkpoint inhibitors such as anti-PD-1 antibodies have shown remarkable clinical success in many different tumor types, the proportion of patients benefiting from this treatment option remains low. Therefore, there is a need to sensitize tumors for immune checkpoint blockade. In this study two approaches were tested, a chemoimmunotherapy approach combining PD-1 checkpoint blockade with doxorubicin (DOX) chemotherapy, and ablation of the sphingosine-1-phosphate (S1P) receptor (S1PR4) based on the following rationale. Chemotherapy was shown to induce immune paralysis which contributes to tumor relapse, while PD-1 signaling was shown to facilitate the acquisition of chemoresistance. Thus, combinatorial chemoimmunotherapy is expected to be beneficial by maintaining or even activating anti-tumor immunity during chemotherapy. S1PR4 is an immune cell specific receptor, whose ablation slowed tumor progression by activating anti-tumor immunity in a mouse model that was previously insensitive to anti-PD-1 monotherapy. This suggested that S1PR4 ablation might pre-activate immunity to sensitize for anti-PD-1 therapy.
To test these combinatorial approaches, two tumor mouse models were employed, namely the MC38 murine adenocarcinoma model as well as the transgenic polyoma middle T oncogene (PyMT) breast cancer model. In the MC38 model, a mild synergistic effect of PD-1 immune checkpoint blockade and S1PR4 ablation was observed, indicated by improved tumor progression and survival as compared to the WT control, and an increased number of tumor-free mice compared to anti-PD-1 therapy alone in WT mice. These observations correlated with an enhanced natural killer (NK) cell infiltrate and increased CXCL9 and CXCL10 production in anti-PD-1 treated S1PR4 KO tumors. As noted before, the PyMT model was largely resistant to anti-PD-1 monotherapy in a therapeutic setting. S1PR4 ablation alone showed significant tumor reduction that was not further enhanced by anti-PD-1 treatment. The same was observed when chemotherapy with DOX was added, where WT tumors relapsed, while S1PR4 KO tumor did not. Addition of anti-PD-1 did only mildly increase tumor control in S1PR4 KO mice, indicating that S1PR4 KO per se very efficiently re-activated anti-tumor immunity. Since S1PR4 KO induces type I 12 interferon (IFN-1) over-production in S1PR4 KO PyMT tumors, a link between high IFN-1 levels and tumor immunity was tested by using mice deficient in the IFN-1 receptor (IFNAR1). Unexpectedly, DOX chemotherapy was most efficient in mice with IFNAR ablation only as compared to WT, S1PR4 KO or S1PR4 and IFNAR1 double KO mice, although deficiency in IFNAR signaling is predominantly regarded as tumor promoting. The underlying mechanisms need to be tested in future studies. Interestingly, chemoimmunotherapy in WT mice prevented tumor relapse to a similar extent than S1PR4 KO and was superior to chemotherapy or immune checkpoint blockade alone. To investigate mechanisms of chemoimmunotherapy success compared to monotherapy, whole transcriptome analysis was used, which identified a set of genes that were upregulated specifically upon chemoimmunotherapy. This gene signature and, more specifically, a condensed four-gene signature predicted favorable survival of human mammary carcinoma patients in the METABRIC cohort.
Moreover, PyMT tumors treated with chemoimmunotherapy contained higher levels of cytotoxic lymphocytes, particularly NK cells. Gene set enrichment analysis and ELISA measurements revealed increased IL-27 production and signaling in PyMT tumors upon chemoimmunotherapy. Moreover, IL-27 improved NK cell cytotoxicity against PyMT cells in vitro. These data supported recent clinical observations indicating a benefit of chemoimmunotherapy compared to monotherapy in breast cancer and suggested potential underlying mechanisms.
Taken together the present work revealed new strategies to reactivate tumor immunity leading to improved chemotherapy response, namely a combination with immune checkpoint blockade and ablation of S1PR4, which activated different lymphocyte compartments within tumors.
Functional roles of COMP and TSP-4 in articular cartilage and their relevance in osteoarthritis
(2020)
Osteoarthritis (OA) is a slowly progressing disease, resulting in the degradation of cartilage and the loss of joint functionality. The cartilage extracellular matrix (ECM) is degraded and undergoes remodelling in OA progression. Chondrocytes start to express degrading proteases but are also reactivated and synthesise ECM proteins. The spectrum of these newly synthesised proteins and their involvement in OA specific processes and cartilage repair is hardly investigated.
Human articular cartilage obtained from OA patients undergoing knee replacement surgery was evaluated according to the OARSI histopathology grading system. Healthy, non-OA cartilage samples were used as controls. The expression and distribution of thrombospondin-4 (TSP-4) and the closely related COMP were analysed on the gene level by PCR and on the protein level by immunohistology and immunoblot assays. The potential of TSP-4 as a diagnostic marker was evaluated by immunoblot assays, using serum samples from OA patients and healthy individuals. The functional role of both proteins was further investigated in in vitro studies using chondrocytes isolated from femoral condyles of healthy pigs. The effect of COMP and TSP-4 on chondrocyte migration and attachment was investigated via transwell and attachment assays, respectively. Moreover, the potential of COMP and TSP-4 to modulate the chondrocyte phenotype by inducing gene expression, ECM protein synthesis and matrix formation was investigated by immunofluorescence staining and qPCR. The activation of cartilage relevant signalling pathways was investigated by immunoblot assays.
These results showed for the first time the presence of TSP-4 in articular cartilage. Its amount dramatically increased in OA compared to healthy cartilage and correlated positively with OA severity. In healthy cartilage TSP-4 was primarily found in the superficial zone while it was wider distributed in the middle and deeper zones of OA cartilage. The amount of specific TSP-4 fragments was increased in sera of OA patients compared to healthy controls, indicating a potential to serve as an OA biomarker. COMP was ubiquitously expressed in healthy cartilage but degraded in early as well as re-expressed in late-stage OA. The overall protein levels between OA severity grades were comparable. Contrary to TSP-4, COMP was localised primarily in the upper zone of OA cartilage, in particular in areas with severe damage. COMP could attract chondrocytes and facilitated their attachment, while TSP-4 did not affect these processes. COMP and TSP 4 were generally weak inducers of gene expression, although both could induce COL2A1 and TSP-4 additionally COL12A1 and ACAN after 6 h. Correlating data were obtained on the protein level: COMP and TSP-4 promoted the synthesis and matrix formation of collagen II, collagen IX, collagen XII and proteoglycans. In parallel, both proteins suppressed chondrocyte hypertrophy and dedifferentiation by reducing collagen X and collagen I. By analysing the effect of COMP and TSP-4 on intracellular signalling, both proteins induced Erk1/2 phosphorylation and TSP-4 could further promote Smad2/3 signalling induced by TGF-β1. None of the two proteins had a direct or modulatory effect on Smad1/5/9 dependent signalling.
In summary, COMP and TSP-4 contribute to ECM maintenance and repair by inducing the expression of essential ECM proteins and suppressing chondrocyte dedifferentiation. These effects might be mediated by Erk1/2 phosphorylation. The presented data demonstrate an important functional role of COMP and TSP-4 in both healthy and OA cartilage and provide a basis for further studies on their potential in clinical applications for OA diagnosis and treatment.
Correct cellular function is ensured by a complex network of proteins and enzymes, regulating protein synthesis and degradation. This protein network, maintaining the so-called protein homeostasis, regulates those processes on multiple levels, producing new or degrading old proteins to cope with changing intra- and extracellular environments. Disturbance of this tightly regulated machinery can have severe effects on the cell and can lead to a variety of pathologies on organism level. Diseases including cancer, neurodegeneration and infections are associated with causative or consequent alterations in protein homeostasis. To understand the pathologies of these diseases, it is therefore critical to examine how perturbations of protein homeostasis affect cellular pathways and physiology. In the recent years, analysis of protein homeostasis networks has resulted in the development of novel therapeutic approaches. However, for many factors it remains unclear how the cell is affected, if they are disturbed. Protein synthesis and degradation represent immediate responses of the cell to changes and need to be studied in the right timeframe, making them difficult to access by common methodology. In this work we developed a new mass spectrometry (MS) based method to study protein synthesis and degradation on a system-wide scale. Multiplexed enhanced protein dynamic (mePROD) MS was developed, overcoming these limitations by special sample mixing and novel data analysis protocols. MePROD thereby enables the measurement of rapid and transient (e.g. minutes) changes in protein synthesis of thousands of proteins. During responses of the cell to stressors (e.g. protein misfolding, oxidation or infection), two major pathways regulate the protein synthesis: the Integrated Stress Response (ISR) and mammalian target of rapamycin (mTOR). Both pathways have been connected with various diseases in the past and are common therapy targets. Although both pathways target protein synthesis in stress responses, the set of targets regulated by these pathways was believed to differ. Through the new mePROD MS method we could measure a comprehensive comparison of both pathways for the first time, revealing comparable system-wide patterns of regulation between the two pathways. This changed the current view on the regulation elicited by these pathways and furthermore represents a useful resource for the whole field of research. We could further develop the mePROD method and decrease MS measurement time needed to obtain an in-depth dataset. Through implementation of logic based instrument methods, it was possible to enhance the number of measured proteins by approximately three-fold within the same measurement time.
The dynamics of protein synthesis and degradation are frequently modulated by pathogens infecting the cell to promote pathogen replication. At the same time, the cell counteracts the infection by modulating protein dynamics as well. To develop useful therapy approaches to fight infections, it therefore is necessary to understand the complex changes within the host cell during infections on a system-wide scale. In 2019, a novel coronavirus spread around the world, causing a world-wide health-crisis. To better understand this novel virus and its infection of the host cell we conducted a study applying the mePROD methodology and classical proteomics to characterize the dynamic changes during the infection course in vitro. We discovered that the infection remodeled a diverse set of host cell pathways (e.g. mRNA splicing, glycolysis, DNA synthesis and protein homeostasis) and thereby showed possible targets for antiviral therapy. By targeted inhibition of these pathways, we could observe that these pathways indeed are necessary for SARS-CoV-2 replication and their inhibition could reduce viral load in the cells. Another experimental approach focused on the dynamic changes of protein modification, namely phosphorylation, after infection with SARS-CoV-2. Here, we could show the very important participation of growth factor signaling pathways in viral proliferation. Both studies together revealed critical pathways that are needed for the viral proliferation and hence are promising candidates for further therapies. Subsequent targeting of these pathways by either already approved drugs (Ribavirin and Sorafenib) or drugs in clinical trials (2-deoxyglucose, Pladienolide-B, NMS-873, Pictilisib, Omipalisib, RO5126766 and Lonafarnib) could block viral replication in vitro and suggests important clinical approaches targeting SARS-COV-2 infection.
Limb stump pain after amputation, due to sensitized neuromas, is a common condition that can cause a great deal of suffering in affected patients. Treatment is difficult, requiring a multidisciplinary approach that is often unsatisfactory. One treatment used to mitigate pain is electrical stimulation (EStim), administered using several different therapeutic approaches. The research described in this dissertation sought to characterize changes in peripheral nerve morphology, and neuroma formation, following limb amputation, with an eye toward developing better treatment strategies, that intervene before neuromas are fully formed. Another focus of this study was to evaluate the effect EStim has on changes in peripheral nerve morphology, and neuroma formation, following limb amputation.
Right forelimbs of 42 male Sprague Dawley rats were amputated. At 3, 7, 28, 60 and 90 days post amputation (DPA) 6 limb stumps, in each group, were harvested and changes in peripheral nerve morphology, and neuroma formation were measured. In addition, limb stumps of 6 EStim treated, 6 sham-treated (deactivated EStim devices), and 6 non-treated rats were harvested at 28 DPA.
Analysis revealed six distinct morphological characteristics of peripheral nerves during nerve regrowth and neuroma development; 1) normal nerve, 2) degenerating axons, 3) axonal sprouts, 4) unorganized bundles of axons in connective tissue, 5) unorganized axon growth into muscles, and 6) unorganized axon growth into fibrotic tissue (neuroma). At the early stages (3 & 7 DPA), normal nerves could be identified throughout the limb stump tissues and small areas of axonal sprouts were present near the distal tip of the stumps. Signs of degenerating axons were evident from 7 to 90 DPA. From day 28 on, variability of nerve characteristics, with signs of unorganized axon growth into muscle and fibrotic tissue, and neuroma formation, became visible in multiple areas of stump tissue. These pathological features became more evident at 60 and 90 DPA. EStim treated stumps revealed neuroma formation in 1 out of 6 animals, whereas in sham and controls, neuroma formation was seen in 4 out of 6 stumps respectively.
We were able to identify 6 separate histological stages of peripheral nerve regrowth and neuroma formation over 90 days following amputation. Axonal regrowth was observed as early as 3 DPA, and signs of unorganized axonal growth and neuroma formation were evident by 28 DPA. Our observations suggest that EStim-based treatment and/or other prevention strategies might be more effective if administered in the initial dynamic stages of neuroma development.
Development of treatment strategies of chronic inflammatory disorders relies on on-going progress in drug discovery approaches and related molecular biologics. This study presents a gene reporter-based approach of phenotypic screening for anti-inflammatory compounds in the context of rheumatoid arthritis (RA).
CEBPD gene, used as the target gene for the screening readout, encodes CCAAT/enhancer binding protein delta (C/EBPδ) transcription factor (TF). Structural and regulatory characteristics of CEBPD gene as well as function of C/EBPδ TF in the context of inflammation satisfied assay requirements. C/EBPδ TF acts as a key regula-tor of inflammatory gene transcription in macrophages (Mϕ) and is observed to con-tribute to disease development in both a rodent model of RA and RA patient biopsies.
Despite well-described pro-inflammatory effects of C/EBPδ TF, it functions as a cell context-specific signal integrator showing also an anti-inflammatory activity. Conse-quently, both activation and inhibition of CEBPD alike may display a desired anti-inflammatory effect. The aim of this study was to develop a high-throughput screening assay for
CEBPD-modulating compounds and confirm hit compounds’ anti-inflammatory effects via gene expression analysis.
Generation and characterization of a multi-gene-reporter cassette 1.0 encoding enzy-matic secreted alkaline phosphatase (SEAP) gene reporter was a priority during the assay development. Chemiluminescent SEAP assay demonstrating high assay sensitivi-ty, broad linear range, high reproducibility and repeatability was chosen to monitor activity of the defined CEBPD promoter (CEBPD::SEAP). PMA-differentiated and M1-polarized THP-1-derived Mϕ stably expressing multi-gene-reporter cassette 1.0 were used as the assay’s cellular system. mRNA expression of both reporter CEBPD::SEAP and endogenous CEBPD mirrored each other in response to a LPS and IFN-g-triggered inflammatory stimulus (M1 treatment), even though the defined CEBPD promoter re-gion, utilized in the assay, contained only the most proximal and known regulatory se-quences. SEAP chemiluminescence in the reporter cells´ supernatant reliably correlat-ed with the M1 treatment-induced CEBPD::SEAP gene expression. The final screening protocol was developed for semi-automatic screening in the 384-well format.
In total, 2054 compounds from LOPAC®1280 and ENZO®774 libraries were screened twice
using the enzymatic SEAP readout with subsequent analysis of 18 selected compounds: nine with the highest and nine with the lowest signals, further characterized by qPCR. Gene expression levels of endogenous CEBPD, CEBPD::SEAP reporter as well as, IL-6,
IL-1β, and CCL2 as inflammatory markers were quantified. qPCR assays failed to corre-late to SEAP readout in 15 compounds within three standard deviations (SDs) from sol-vent control: nine low signal and six high signal compounds. Demonstrating both assay sensitivity and specificity, a correlation between qPCR gene expression and SEAP readout was observed for three hit compounds with signals above three SDs: BET inhib-itors (BETi) GSK 1210151A and Ro 11-1464 as well as an HDAC inhibitor (HDACi) vori-nostat. The control compound trichostatin A (TSA) that reproducibly upregulated SEAP readout is also an HDAC inhibitor with a similar structure to vorinostat and was there-fore included in the anti-inflammatory phenotype analysis.
The observed suppression of IL-6, IL-1ß, and CCL2 gene expression by hit compounds suggested their anti-inflammatory effect in THP-1 reporter Mϕ. mRNA expression of
IL-6 and CCL2 was suppressed by HDACi and BETi at both 4 and 24 hours, while BETi reduced IL-1β mRNA expression 24 hour time point. BETi significantly upregulated gene expression of both reporter CEBPD::SEAP and endogenous CEBPD, 4 hours after M1 treatment. At the same time point, HDACi completely abolished the mRNA expres-sion of the endogenous CEBPD, while simultaneously upregulating mRNA expression of the reporter CEBPD::SEAP. The use of the most proximal 300 base pairs region of en-dogenous CEBPD promoter, making the upstream regulatory elements unavailable in the assay, may account for differential expression levels of SEAP and C/EBPδ TF. This observation corroborated the need to include a longer and more extensive CEBPD´s gene regulatory area. Thus, an improved multi-gene-reporter cassette 2.0 was gener-ated to be used on the basis of a bacterial artificial chromosome (BAC) covering CE-BPD´s genomic area of about 200,000 base pairs.
The generated screening assay is flexible, reliable, and sensitive displaying potential for drug discovery and drug repurposing. The pharmacological modulation of CEBPD gene expression, first reported for GSK 1210151A, Ro 11-1464, and vorinostat, contrib-utes to the understanding of inflammatory responses in Mϕ and may have RA thera-peutic applications.