Qatar Foundation Annual Research Conference Proceedings Volume 2016 Issue 1

Neuroblastoma, a type of solid malignant tumour diagnosed during infancy represents around 10% of all paediatric cancers, marking it as the second most common paediatric cancer. It is identified as a highly heterogeneous tumour that varies from persistent progression to a spontaneous progression. Development of neuroblast masses occur mostly in abdomen (65%), chest (20%), neck (5%) or pelvis (5%) and is classified to four major stages as L1 (very low), L2 (low), M (intermediate) and MS (risk) group of patients. Intracellular calcium ([Ca²⁺]i), the secondary messenger, plays a vital role in regulating cellular processes and hence maintaining the cellular calcium homeostasis is critical. Levels of [Ca²⁺]i in cancer cells is eminent as it modulates the proliferative or apoptotic pathway of the cell, bestowing the effect of anti-cancer drugs on [Ca²⁺]i. Here we focus on the effect of [Ca²⁺]i in the development and treatment of neuroblastoma. In resting cells, the concentration of [Ca²⁺]i ranges between 10 and 100 nM, reaching up to a 100 fold increase upon the Ca²⁺ entry into the cells from the extracellular space or its release from the internal Ca²⁺ stores (endoplasmic reticulum and mitochondria). [Ca²⁺]i homeostasis is maintained in the cell by a Ca²⁺ influx and efflux mechanism, in which the inositol-1,4,5 triphosphate receptor (InsP3R) and the ryanodine receptor (RYR) play an important role in regulating the influx mechanism. Signalling pathways involved in association to neuroblastoma includes growth factors like Epidermal Growth factor (EGF), Insulin-like Growth Factor (IGF), Nerve Growth factor (NGF), Platelet-derived Growth Factor (PDGF) and Vascular Endothelial Growth Factor (VEGF). Activation of these growth factor signalling pathways, activates a cascade of downstream signalling molecules including PI3K/AKT, ALK and FAK as intermediate kinases to MYCN, NF-KB and p53 as important transcription factors involved in the development of neuroblastoma. [Ca²⁺]i and PI3K/AKT pathway interactions lead to a loop of continuous activation of this cascade, while the activation of ALK and FAK leads to the activation of calmodulins and CaM dependent protein kinase kinase. Several studies confers the role of [Ca²⁺]i in inducing differentiation, proliferation and apoptosis in neuroblastoma. This poster explains in detail on the intracellular pathways that regulate differentiation, proliferation and apoptosis in neuroblastoma and the mechanism by which the [Ca²⁺]i homeostasis is maintained in the cells. Calcium-Sensing Receptor (CaSR), a G-protein coupled receptor is a vital mediator protein that sustains the cellular responses and determines the cell fate in response to the external Ca²⁺ concentration between 0.05-5 mM from a proliferative stage to a stage of quiescence and differentiation. CaSR activation is often associated with an up regulated expression of parathyroid hormone related peptide (PTHrP), which has a role in inducing hyperglycemia in cancer cells. It is conferred that these CaSR exerts tumour suppressor functions in neuroblastoma and is found in differentiated favourable neuroblastoma tumours. Evidences suggest that in neuroblastoma cells, common chemotherapeutic drugs like arsenic trioxide (As2O3), cisplatin (CDDP) and trimethytin chloride (TMT) that are used in treatment of cancer interferes with the [Ca²⁺]i homeostasis and induce apoptosis. It suggests a promising role for targeting [Ca²⁺]i for the development of a new drug. Thus, in a nut shell, we elucidate the intracellular mechanisms that are associated with the development of neuroblastoma with the key focus on the [Ca²⁺]i along with possibility of development of a new drug target in the treatment of neuroblastoma.

Background

Technological breakthroughs witnessed over the past decade have led to an explosive increase in molecular profiling capabilities. This has ushered a new “data-rich era” for biomedical researchers. Indeed the recent availability of vast compendia of biomedical “Big Data” offers unique opportunities to devise novel approaches to knowledge discovery. We have launched an innovative “Collective Data to Knowledge” (CD2K) platform at the Sidra Medical and Research Center, which provides a hands-on accelerated training path for young biomedical researchers. The originality of this approach stems from the fact that it does not rely on de novo generation of data but leverages instead large dataset collections available in public repositories for the discovery of novel scientific knowledge. It does however recapitulates all other steps involved on the path to transformation of data into novel biomedical knowledge, including knowledge gap assessment and prioritization, hypothesis generation and testing, and generation of reports for publication in peer reviewed journals. Furthermore, besides providing accelerated hands-on training it also potentially constitutes a highly efficient approach to the generation of intellectual capital in Qatar.

Methods

The approach that we have devised relies on a wide range of available bioinformatics tools and resources.

a) Data integration and dissemination: For data dissemination we rely on a custom web application – the gene expression browser – that is used for integration of heterogeneous data (e.g. molecular profiles, together with clinical information, sample information and results from ancillary assays) [1]. It provides user with seamless access to large and complex datasets that can be viewed in an interactive format. This tool has now been deployed at Sidra Medical and Research Center and is used to create curated themed dataset collections that will be described in peer-reviewed communications (manuscripts in preparation).

b) Knowledge gap assessment: Knowledge gaps are identified via profiling of the biomedical literature for sets of differentially expressed genes. For instance knowledge gaps may be revealed among the hundreds of genes identified via transcriptome profiling as being induced by TNF-α, a host-derived pro-inflammatory cytokine. Among those genes many will be associated in the literature with pro-inflammatory responses, but a number of them will be shown via literature profiling not to be associated with inflammation, thus constituting a knowledge gap in which lies the opportunity for discovery.

c) Hypothesis generation and In Silico validation: the identification of knowledge gaps is the first step towards generation of novel knowledge. Next we rely on tens of thousands of publically available datasets to validate and extend the initial finding, often leading to formulation of novel hypotheses that can be immediately tested by accessing other relevant datasets.

d) Information extraction: We also devised standardized approaches for extracting and structuring information. These methods are used when profiling the literature and dataset collections in view of preparing the background and result sections of the reports. This principled approach helps trainees with manuscript preparation, which often constitutes a hurdle for scientists at an early in their careers.

e) Knowledge dissemination: trainees are then encouraged to submit their work in peer-reviewed scientific journals. It is the opportunity for them to learn about this essential process that is one of the cornerstones of the scientific discovery process.

Results

Workshops carried out in Qatar and in several countries around the world have been instrumental in the development of a CD2K training curriculum. In addition, a proof of principle of the effectiveness of the CD2K platform has been established with identification in a short amount of time of new discoveries with potential for high impact.

1) CD2K Training Workshops

We have conducted hands-on training workshops this year for 6 organizations. Each workshop spanned between 1 to 3 days and involving overall more than 100 participants.

An introductory CD2K training workshop was organized at the Sidra Medical and Research Center. Training material was further developed by conducting CD2K workshops in a wide range of settings: in academic research institutes, in the United States at the Jackson Laboratory for Genomics Medicine and in Singapore at the A-star Institute; in a University in Thailand (Chulalongkorn University, Bangkok); in a research hospital in France (Hopital Europeen, Marseille); in a large pharmaceutical company in the United States (MedImmune, a subsidiary of AstraZenca, in Gaithersburg, Maryland).

These workshops were instrumental to the establishment of a robust training curriculum; consisting in the following learning objectives:

a) Collective biomedical data profiling

b) Literature profiling

c) Identification and prioritization of knowledge gaps

d) Hypothesis generation and in silico validation

e) Information extraction

f) Knowledge dissemination

2) CD2K Proof of principle

A post-doctoral research fellow has been assigned to the piloting the CD2K platform at Sidra, with the objective of identifying and prioritizing potential knowledge gaps and submitting reports for publication in peer reviewed journals within 12 months for three novel findings with high potential for translation. At the time of submission of this abstract 10 months within this pilot 2 manuscripts have been submitted with a third one being finalized. The first two articles have appeared online pre-peer review in March and September of this year in the journal “Faculty of 1000 Research”:

a) The first article reports the identification of “ADAM metallopeptidase 9” (ADAM9) as a candidate biomarker for the specific assessment of tissue damage caused by infection, independently of pathogen-driven inflammatory processes [2]. This work revealed a new potential role for ADAM9 in immunological homeostasis and pathogenesis. The abundance of ADAM9 transcripts in the blood was increased in patients with acute infection but changed very little after in vitro exposure to a wide range of pathogen-associated molecular patterns (PAMPs). Furthermore it was found to increase significantly in subjects as a result of tissue injury or tissue remodeling, in absence of infectious processes. Therefore this marker could potentially be used as a triage tool for patients presenting with symptoms of infection in the emergency room that may or may not require hospitalization

b) The second article reported the identification of blood molecular signatures that correlate with protection, or lack thereof, conferred by the RTS,S malaria vaccine [3]. This finding is important because this vaccine, which was licensed this year by European regulatory authorities, only protects about 40% of vaccinated individuals. Understanding the mechanisms that undermine the efficacy of this vaccine could lead to the development of a universally protective prophylactic modality for a disease that affects about 200 millions people and causes 500,000 deaths each year worldwide.

c) A third report that is being finalized investigates the role of Aquaporin 9 (AQP9), a water-selective membrane channel protein. This molecule is regulated during infection and appears to play a role in maintaining elevated metabolism associated with inflammatory responses. It may also inadvertently promote pathogen growth with adverse consequences in conditions such as pregnancy where elevated baseline metabolic states may contribute to enhance disease severity. Finally our observations also suggest a role for AQP9 in mediating pathogen clearance via phagocytosis.

Conclusions

The approach that we have devised recapitulates all the steps involved in the scientific discovery process, from data interpretation to knowledge dissemination. It allows screening, identification and prioritization of potential knowledge gaps, followed by in sillico validation and hypothesis generation and testing, finally resulting in preparation and publication of reports in a peer-reviewed journal. Its effectiveness as a training platform stems from the fact that it does not rely on de novo generation of data for discovery and validation. It leverages instead the vast amounts of available biomedical data, which will allow for accelerated and highly efficient hands-on training of aspiring biomedical researchers.

References

[1] Speake C, Presnell S, Domico K, Zeitner B, Bjork A, Anderson D et al. An interactive web application for the dissemination of human systems immunology data. J Transl Med. 2015;13:196. doi:10.1186/s12967-015-0541-x

[2] Rinchai D, Kewcharoenwong C, Kessler B et al. Abundance of ADAM9 transcripts increases in the blood in response to tissue damage [version 1; referees: 3 approved with reservations] F1000Research 2015, 4:89 (doi: 10.12688/f1000research.6241.1)

[3] Rinchai D, Presnell S and Chaussabel D. Blood Interferon Signatures Putatively Link Lack of Protection Conferred by the RTS,S Recombinant Malaria Vaccine to an Antigen-specific IgE Response [version 1; referees: awaiting peer review] F1000Research 2015, 4:919 (doi: 10.12688/f1000research.7093.1)

Understanding population level food consumption, which has considerable influence on population health, is a major challenge. For instance, obesity, which is particularly pressing in the Gulf region, is largely driven by changes in food consumption. Many other widespread diseases, such as diabetes, heart disease, and high blood pressure, are directly or indirectly affected by eating habits. Understanding food consumption generally involves surveys and self reporting which introduce certain biases, latency and substantial cost. Social media offers new possibilities to passively monitor and study eating behaviors, and track them in real-time across regions and time.

Predicting population level statistics (e.g. tracking seasonal epidemics like Flu) can be obtained through social media (e.g. shared tags and texts) which provides large scale, non-intrusive, and location-aware (regional) data in real time. Instagram is a hugely popular image sharing application, particularly in the Gulf region. Although users often annotate their social media posts with hashtags, a lot more information remains “hidden” in the actual image, requiring novel processing methods. Noting that “a picture is worth a thousand words”, we make use of this visual information through state-of-the-art deep learning models, particularly concentrated on food-related images.

We propose a method for tracking regional and temporal food habits though social media images. Concerning technical aspects, we have two major contributions: (a) learning visual concepts from social media images with extremely noisy labels, and (b) predicting regional statistics through visual information analysis.

The recent developments in scene and image categorization, particularly the advances in deep learning, show that with large quantities of data (i.e. big data) we can achieve great performance, approaching the level of human annotations. Here we show that even with extremely incomplete labels (i.e. only a limited set of tags exist for images taken from instagram) a robust auto-tagging system can be learned using deep convolutional neural networks, particularly with the help of millions of images. We mainly focus on food images and food related tags collected from instagram. We explored two major deep learning architectures for food label prediction: the Alexnet [1] (see Fig. 1) and VGG network[2].

We explored both training from scratch (i.e. learning the system only using instagram images), and finetuning an existing convolutional network (i.e. build upon an already trained system for object categorization using large-scale imagenet database). Although both models have very strong prediction capabilities, VGG network gives better predictions overall. Figure 1 shows some example images auto-tagged with food labels using our system.

We employ our food label predictor for tracking regional and temporal food habits. These population level statistics not only inform us on healthy eating habits but also provide us cues for predicting consequences of these behaviors such as conditions and diseases. We show that certain public health statistics (e.g. alcohol consumption) can be reliably predicted by analyzing social media images through our deep learning models.

The data collection and fusion is another major issue, particularly required for real-time analysis. We developed a system that can reliably identify all the geo-location tags (e.g. places that are used while tagging the location of the images - houses, cafes, etc.) by performing a large scale grid search using geographic tessellation models. We also developed a distributed system that can keep track of the shared images in the identified locations in real time. In depth analysis are performed over these collected images.

One of the potential outcomes of our project is a system that can visualize regional and temporal food habits which can be used as a tool for predicting the future habits and understanding main dynamics that affect the food consumption. The system can track the food habits across regions, cultures and time (e.g. daily, seasonal, yearly). For instance it will be possible to see how fast food is gaining or losing prominence compared to healthier food, and in which regions it happens more drastically. We can track Christmas dinner changes over time, or what different regions of the world ate for Christmas. We can display what type of dishes are mostly consumed in the month of Ramadan, and if food image sharing is changed during the day and the night. Many other potential applications can be build upon the proposed system.

References

[1] Krizhevsky, A., Sutskever, I., and Hinton, G. E. ImageNet classification with deep convolutional neural networks. In NIPS, pp. 1106–1114, 2012.

[2] K. Simonyan and A. Zisserman. Very deep convolutional networks for large-scale image recognition. CoRR, abs/1409.1556, 2014.

Background

Respiratory viruses have a predictable seasonality, which varies regionally. The reason for such seasonality is not well known yet, but atmospheric factors such as high humidity and temperature may assist virus survival in small particle droplets or aerosols, and on infected surfaces.

Objective

The goal of the study was to determine the seasonal variation in respiratory syncytial virus (RSV) infection in a desert climate.

Methods

A retrospective and cross sectional study was performed at Hamad Medical Corporation (HMC), the only tertiary and academic medical center in the State of Qatar. The study included infants and young children ages 0 to 24 months that were admitted to our pediatric ward with diagnosis of acute bronchiolitis from the period of January 2010 to December 2012. The following information were collected: gestational age, gender, respiratory virus real time polymerase chain reaction (RVRT-PCR) conducted on nasopharyngeal secretions, and hospital length of stay (LOS).

Results

835 infants and young children met the study criteria with mean age at diagnosis of 3.61 ± 3.56 months ranging from 0.33 to 24 months. RVRT-PCR was performed on 769 (92.0%) of the participants. RSV was positive in 352 (45.7%) children admitted with clinical bronchiolitis. In addition, no viruses were identified in 142 (18.4%), and respiratory viruses other RSV were found in 275 (35.7F%) of children. Our investigation shows that there has been a steady and periodic seasonal variation in the RSV rate over the study period. A seasonal trend for the RSV (detected by RVRT-PCR) rate was evident (Fig. 1), showing annual peaks in the months of October, November, December, and January, with a significant test for seasonality (test statistics [T] =  3.15, P = 0.009).

Conclusions

In countries with desert hot weather, bronchiolitis might affect infants and children throughout the year. Our results suggest that the combination of uninterrupted RSV seasonality can provide factual guidance for healthcare planning and application of RSV prevention scheme, such as extending the palivizumab vaccine series.

Figure 1: Sequence chart for RSV rate of infection during various months in children admitted with acute clinical bronchiolitis

Summary

In this paper, we present a novel interpolation-based Stokes image reconstruction scheme for the division-of-focal-plane (DoFP) polarization image sensors. Different from the previous implementations, our proposed method first demosaics the raw image by mainstream interpolation algorithms then converts the up-sampled images to Stokes images with much richer polarization-related physical information. This not only leads to significant resolution improvement to the captured raw image, but also greatly reduces the caused pixel mean square error (MSE). Experimental data from the test images have validated the effectiveness of this proposed scheme.

Motivation

Solid-state image sensors, which are capable of extracting the incident light's polarization information in addition to intensity and color (i.e. wavelength), take great advantages in a wide range of applications [1]. By looking through a layer of patterned micrometer-scale pixelated polarizing elements, a set of mosaicked polarization raw sub-images can be generated simultaneously, namely DoFP polarization imaging. As shown in Fig. 1 (a), similar to the widely-exploited Bayer pattern of color imaging, the mosaicked polarization raw sub-images down-sample each polarization channel by 75%, leading to significant spatial resolution loss. Meanwhile, in order to make the raw sub-images physically meaningful, they are typically translated to Stokes sub-images, which correspond to first three Stokes parameters representing the unpolarized and linearly polarized components of the incident light. In the previously reported implementations, the neighboring four sub-pixels (i.e. I0, I90, I45, I135) are directly substituted to the Stokes parameters' classical expressions. As a result, a 75% down-sampled Stokes images are acquired. In order to compensate this resolution loss, we propose a new Stokes image reconstruction scheme: before the Stokes image conversion, interpolate the mosaicked polarization raw sub-images with mainstream algorithms, including linear, cubic and spline.

Results

Figure 1 (c) illustrates the proposed detailed image reconstruction flow. In addition, we have also compared our proposed method to the traditional implementation of directly applying the same interpolation algorithm to the aforesaid 75% down-sampled Stokes images [Fig. 1 (b)]. For the extracted down-sampled Stokes images in the previous implementations, even with the same interpolation algorithms applied, our proposed method still outperforms by almost 10 folds in term of Stokes parameter S2's MSE (Fig. 2), which well-balances the spatial resolution compensation and the Stokes image generation by minimizing the caused overall MSE. Figure 3 illustrates the real images with different Stokes image reconstruction schemes applied.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Grant No. 61504087), the Kongque Technology Innovation Foundation of Shenzhen (Grant No. KQCX20120807153227588), the Fundamental Research Foundation of Shenzhen (Grant No. JCYJ20140418095735624, and JCYJ20150324141711677).

Reference

[1] M. Kulkarni, V. Gruev, “Integrated spectral-polarization imaging sensor with aluminum nanowire polarization filters”, Optics Express, vol. 20, no. 21, pp. 22997–23012, 2012.

1. Background

Stroke, i.e. loss of brain function(s) due to disturbance in the blood supply to the brain, is the main cause of adult disability (e.g. paralysis) in the world, leaving more than half of the patients dependent on daily assistance. In Qatar, stroke is a major health problem with an estimated incidence of 238/100,000 per year for the population over 45 years old [1]. Stroke patients are often hospitalized and/or subjected to intensive rehabilitation programs for long periods of time, and their quality of life is severely affected socially and economically. Around 10% of the hospital beds in Qatar are occupied by stroke patients [1]. Thus, without major advances to improve prevention, treatment and rehabilitation of stroke, the social and economic costs of this disease will increase dramatically.

There are pathological and physiological changes on the cellular and molecular levels associated with stroke. The objective of this work is to determine the molecular and structural changes occurring in the tissue of rat's brain. Vibrational spectroscopy, i.e. Fourier transform infrared (FTIR) imaging spectroscopy, was used as rapid and objective diagnostic platform to investigate the pathological and pathological changes in the rat's brain sections three weeks after stroke. FTIR spectroscopy was also used to differentiate between the biochemical makeup of the white and grey matters of a healthy control brain samples. Also, in the current study, scanning electron (SEM), energy dispersive X-ray spectroscopy (EDX), and atomic force microscopic (AFM) techniques were assessed to study the structural changes in the rat's brain tissues after experiencing an induced stroke.

2. Experimental

2.1. Sample preparation

Rats were anesthetized using 2–3% isoflurane. Experimental stroke was induced in rats by 90-min occlusion of the right middle cerebral artery with an intraluminal filament. Rats were euthanized with a lethal dose of sodium pentobarbital and transcardially perfused with 4% paraformaldehyde. Rat's brains were extracted, embedded in paraffin and then serially sliced, using semi-automated rotary microtome, into 5 μm thickness sections for the FTIR imaging and AFM analysis and 35 μm thickness for the SEM and EDX analysis. The brain sections were mounted on MirrIR CFR, Low-e microscope slides for the FTIR imaging analysis, and on aluminum metal for the SEM analysis and EDX analysis. The paraffin was removed from the samples by using xylene and isopropanol.

2.2. Instrumentation

2.2.1. FTIR Imaging Measurements

The FTIR images were obtained using FTIR spectrometer (Agilent Technology) at a reflection mode within the range of 4000–700 cm^–1. Spectral images were analyzed using Metlab software (The Mathworks Inc.). Origin 2015 software was used for graph drawing. Principal component analysis (PCA) was performed to study the spectral data variations between the FTIR spectra and images.

2.2.2. Scanning Electron Microscopy (SEM)

Rat's brain sections of 35 μm thickness were mounted on aluminum slides for SEM analysis. All the samples were viewed with a FEI Quanta 200, USA scanning electron microscope at 10 kV. SEM micrographs of the brain stroke and healthy rat's sections were compared. Elemental distribution in both healthy and induced stroke brain sections were investigated by using energy dispersive X-ray spectroscopy (EDX) equipped with SEM. The spectra provided a semi-quantitative view of the elemental composition of both weight and atomic percent.

2.2.3. Atomic Force Microscopy (AFM)

Bruker atomic force microscopy (AFM) was used for imaging and quantitatively determining the local elastic properties of healthy and induced stroke rat's brain sections. A controllable and constant force was applied at each data point and using the resulting force-distant curve for the formation the AFM images. Brain sections were scanned at 10 μm by 10 μm. About 100 force-distance curve were collected for each healthy and induced stroke brain sections and two random scan lines of force-distance curves was recorded.

3. Results and Discussions

The FTIR spectroscopy results indicated that the white matter is richer in lipid content than the grey matter as shown in Figs. 1 and 2. The infrared spectrum images showed a decrease in the lipid content of the white matter associated with the induced stroke brain sections. FTIR bands assigned to the bio-chemical makeup such as proteins, lipids and ester varied in positions, line-shape, and intensity between control and induced stroke brain samples. The spectral images showed that there is a configuration changes is associated with the lipid bands in the rat's brain white matter that experienced stroke.

The FTIR spectral images of the white matter in the induced stroke brain sections indicated that amide I and ester bands experienced a bio-chemical changes as shown in Fig. 3 and 4. Figure 5 shows the second derivative of the collected FTIR spectra from induced stroke brain sections. In Fig. 5, there are spectral differences that assigned to ester and protein regions. Figure 6a represents the loading spectra of the first three principal component analysis (PC1, PC2 and PC3). The variations principally were located in the regions of amide I band at (∼1695-1637 cm^–1) and small variation in the amide II band at (1543 cm^–1). Figure 6b represents the loading spectra of the PC4, PC5 and PC6. The variations principally were located in the protein region, mainly amide I band at (∼1695-1637 cm^–1) and ester band at about 1730 cm^–1. The use of FTIR imaging and chemometric analyses such as principal component analysis (PCA) of spectral data allows to investigate and differentiate spectral images pattern collected from control and stroke rat's brain samples.

The scanning electron microscope results showed that lesion region in the induced stroke brain sections are enriched by the selected elements such as Fe and Ca as shown in Fig. 7 (a & b). Scanning electron microscope (SEM) micrographs indicated that there is structure change in the induced stroke brain section. The structure of stroked brain sample in the nanometer scale appeared to be significantly rough compared to the control brain sample (Fig. 8 a & b).

Atomic force microscope (AFM) images showed that the stroke brain section is swollen compared to healthy brain sections. The AFM images of the induced stroke brain sections appeared more stretched when compared to the control brain section image as shown in Fig. 9 (a & b). AFM results also showed that the force-distance curves in Fig. 10, recorded using control (healthy) brain sections (blue) and induced stroke brain sections (red). The force-distance showed that the AFM cantilelver deflection of the healthy brain samples is higher than the induced stroke brain section. This indicate that the healthy brain section are softer and elastic than the induced stroke brain sections.

4. Conclusion

FTIR imaging spectroscopy, scanning electron and atomic force microscopy techniques were able to analyze and differentiate between the healthy and induced stroke rat's brain sections on the molecular, structural and global levels making them valuable tools to investigate, diagnose and study the structural plasticity of the stroke induced brain. FTIR imaging spectroscopy in combination with multivariate analysis such as principal component analysis (PCA) is a non-destructive technique that proves to be rapid, accurate and straightforward to be performed. It constitutes a powerful approach to be used as a medical diagnosis tool to investigate the pathological changes associated with stroke in the brain tissues.

Keywords

Fourier Transform Infrared (FTIR) imaging spectroscopy, Scanning Electron Microscopy (SEM), Atomic Force Microscopy (AFM), Brain tissue, Stroke, Chemometric Analysis

Acknowledgments

This article was made possible by a NPRP award [5 - 381 - 3 - 10] from the Qatar National Research Fund (a member of The Qatar Foundation). The statements made herein are solely the responsibility of the authors. We would like to acknowledge the Center for Advanced Materials at Qatar University for performing the AFM analysis, and Central Laboratories Unit at Qatar University for performing the SEM and EDX analysis.

Reference

[1] Hamad, A., Sokrab, T.E., Momeni, S., Mesraoua, B., and Lingren, A., Stroke in Qatar: a one-year, hospital-based study. J Stroke Cerebrovasc Dis, 2001. 10(5): p. 236–41.

Thymidylate kinases (TMKs) play a central role in the production of nucleotide precursors that are required for the replication of DNA. Consequently, this enzyme is a potential drug target for the discovery of anti-bacterial, anti-fungal, and anti-parasitic drugs. In addition, TMKs are also involved in the activation of prodrugs. In particular, the anti-HIV drug AZT is activated by human TMK (huTMK) and the low efficiency of huTMK towards AZT is a significant problem in the use of AZT in the treatment of HIV. Finally, nucleotide precursors are required in large amounts by cancerous cells, thus the inhibition of huTMK by chemotherapeutic agents may enhanced the arsenal of drugs that are used to treat cancer.

Although there has been some effort to develop inhibitors of TMKs, these efforts have been hampered by the difficulty in performing high throughput screening using compound libraries. In addition, the characterization of TMK-drug complexes has been limited to X-ray diffraction studies which provide static information about the enzyme-drug complex. There have been no attempts to apply high-resolution multi-nuclear NMR techniques to determine the fundamental dynamic properties of these enzymes and how the structure and dynamics of the enzyme are altered by the binding of substrates or inhibitors.

As a preliminary step in characterizing these enzymes by NMR we have over-expressed TMKs from yeast, human, and two pathogens - Plasmodium falciparum and Candida albicans. Expression of these TMKs was optimized by the design of synthetic genes for expression in bacteria. In the case of the human enzyme, we are able to routinely produce 250 mg of the enzyme/L of culture. Preliminary NMR spectra of the yeast, human, and plasmodium enzyme show that the protein is a homo-dimer in solution, as anticipated from X-ray studies. The amide and methyl spectra are well resolved, indicating that resonance assignment by traditional TROSY based methods will be feasible for both the amides and the methyl resonances. In particular the high sensitivity and dispersion of the methyl spectra will facilitate characterization of the dynamic properties of these enzymes by carbon and deuterium relaxation. Ligand induced changes in the dynamics and structure of huTMK in solution will be characterized using NMR methods. These studies will provide additional insights into the inability to huTMK to effectively activate AZT. The entropic component of the thermodynamics of substrate binding to TMK from the parasite that causes malaria will also be characterized by determining dynamic changes by NMR methods.

The development of NMR methods to study these enzymes also provides a method for high throughput screening of compound libraries by detecting chemical shift changes in the NMR spectral of the enzyme due to binding of a potential lead compound.

Background

Lung cancer is the second most common cancer in both men and women and it is the leading cause of cancer deaths worldwide. Lung cancer can be divided into two broad categories: non-small cell lung cancer (NSCLC), which consists of about 85% of all lung cancers and small cell lung cancer (SCLC), which account for 15% of all lung cancers. The evolution of lung cancer is a multistep process involving genetic and epigenetic alterations. Standard treatment therapies such as radiation therapy, chemotherapy and surgery has reached a plateau phase. As a result, much work has centered on identifying the molecular targets involved in the tumor cell proliferation, survival and metastasis in effort to identify novel therapeutic approaches. p90 ribosomal S6 kinase (p90RSK) constitutes a family of serine/threonine kinases that have been shown to be involved in cell proliferation of various malignancies via direct or indirect effects on the cell-cycle machinery.

Objectives

To investigate the role of p90RSK in lung adenocarcinomas and whether the inhibition of p90RSK diminishes cancer progression. Moreover, we investigated the involvement of glycogen synthase kinase-3β (GSK-3β) and osteopontin (OPN) in the p90 RSK induced lung adenocarcinoma progression.

Methods

p90RSK, OPN, GSK-3β protein expression were examined in the A549 human lung adenocarcinoma cell line in the presence and absence of BI-D1870 (BID), a p90RSK inhibitor. Gene expression of anti-apoptotic and pro-apoptotic markers namely Bcl2 and Bax, respectively, were studied by reverse transcription polymerase chain reaction. In addition, the A549 lung adenocarcinoma cell line was characterized for cell proliferation using the MTT assay and cell migration using the scratch migration assay.

Results

Our study revealed that the treatment of the A549 lung adenocarcinoma cell line with BID resulted in a significant reduction in protein expression of p90RSK 1 (69.32 ±  12.41% of control; P <  0.05). The inhibition of p90RSK also showed a significant suppression of cell proliferation (54.3 ±  6.73% of control; P

Breast cancer is the most common type of tumor in women in the MENA region and it represents about 20% of the cancers diagnosed every year in Qatar. Although the implementation of cancer therapy has led to an improvement of patients' survival, metastatic breast cancer remains an incurable condition. Immunotherapy is emerging as an innovative therapeutic tool able to cure in some cases established metastatic tumors such as melanoma. There is a growing interest in exploring this fascinating approach in breast cancer. However, little is known about the immune biology of this aggressive disease. Studies from our and other groups have described intratumoral immune gene signatures associated with better responsiveness to immunotherapy and prolonged survival (Galon, Angell, Bedognetti and Marincola, Immunity 2013; PMID: 23890060). In general, these gene signatures imply the activation of interferon stimulated genes (e.g. IRF1 and STAT1), the recruitment of lymphocytes through CXCR3/CCR5 ligand chemokines, and the activation of immune-effector functions such as PRF1 and GZM1 (Wang, Bedognetti, Marincola, JCO 2013; PMID: 23715576). We refer to these genes as the Immunologic-Constant-of-Rejection (ICR) (Bedognetti, Wang, Sertoli, Marincola, Exp Rev Vacc, 2010; PMID: 20518712). The activation of the ICR pathways is accompanied by the counter-activation of immune-regulatory mechanisms such as the expression of PD-L1 and IDO1. Based on these observations, we hypothesize that tumors can be divided in two opposite phenotypes (Bedognetti, Hendrickx, Marincola, Miller, Curr Opin Onc 2015; PMID: 26418235). The first phenotype displays the activation of pro-inflammatory (ICR) and immune-regulatory mechanisms and is characterized by a favorable prognosis and responsiveness to immune manipulations. The second phenotype lacks these two characteristics and is associated by a poor prognosis and resistance to immune manipulations. Whether the development of such different immune phenotypes is influenced by the intrinsic genetics of the tumor cells is presently unclear. The understanding of genetic mechanisms associated with differential immune response can lead the development of targeted approaches. To answer this critical question we analyzed copy number variation, exome and RNA sequencing data from the The Cancer Genome Atlas (TCGA) breast cancer dataset. By using consensus clustering analysis based on RNA-seq data of more than 1000 breast cancer samples, we defined four immunophenotypes characterized by progressive expression of the ICR genes (ICR1