Tissue engineering of any tissue type requires the combination of a supporting scaffold, a range of biological factors and a suitable source of cells. This source of cells must satisfy a number of criteria:- • The ability to form all of the mature/specialised cell types found in the target tissue. • Readily obtainable. • Readily maintainable in the laboratory without requiring excessive resources or time. For intestinal tissue engineering there are a number of issues associated with the use of tissue derived stem cells particularly quantity of normal tissue available (from the patient) and maintenance and expansion of the cells when cultured in vitro. Using embryonic stem cells offers a potential alternative strategy but methods must be established to efficiently differentiate the cells towards the desired fate. Many strategies for differentiating embryonic stem cells are based upon treatment with growth factors in vitro. There is a (logistical) limit to the degree of complexity that these systems can achieve and therefore a limit to the number of differentiation signals that occur during in vivo development that can be mimicked. In recent years using embryonic tissue to provide signals to undifferentiated cells has proved a successful method of directing the differentiation of naïve cells towards a particular fate (with the choice of tissue determined by the desired target cell type). The aims of this thesis were to explore the potential of differentiating embryonic stem cells towards the intestinal progenitor fate using a combination of in vitro cell culture treatment with the growth factor Activin-A and ex vivo co-culture with embryonic chick gut tissue. Previous studies [Kubo et al 2004, Tada et al 2005, Yasunaga et al 2005, D’Amour et al 2005, MacClean et al 2007] have shown that Activin-A treatment will induce embryonic cells to more efficiently differentiate to definitive endoderm, the germ layer from which the intestines (and other visceral organs) arise. These techniques were applied to the Columnar Epithelial Epiblast murine embryonic stem cell line and cell differentiation was then evaluated at the molecular level using Reverse Transcription-Polymerase Chain Reaction, immunocytochemistry and Western blotting. Activin-A treatment produced an upregulation of definitive endoderm markers at both the mRNA and proteomic levels compared to the control conditions. However the cell population produced retained expression of pluripotent markers and showed some expression of markers of other cell lineages. Further studies [Sugie et al 2005, Fair et al 2003, Van Vranken et al 2005, Coleman et al 2007, Krassowska et al 2006] have shown that co-culture of embryonic stem cells with early stage embryonic tissue can induce the formation of particular tissue types; the tissue must be selected based on proximity to the target cell type during development. This exposes the embryonic stem cells to the signals that prompt differentiation towards the target tissue during normal development. With gut tissue much signalling occurs between the different tissue layers that make up the whole organ both during development and in adult tissue. Ex vivo co-culture of murine embryonic stem cells with embryonic chick gut tissue was used to direct their differentiation to the intestinal epithelial stem cell fate. Before the co-culture was carried out various experiments were carried out to establish if the proposed protocol was viable e.g. defining how long chick gut tissue explants could survive in culture. Once this had been established co-culture experiments were undertaken and cell differentiation was then evaluated at the molecular level using Reverse Transcription-Polymerase Chain Reaction, immunocytochemistry and Western blotting. The cells showed some expression of intestinal epithelial stem cell markers at both the mRNA and proteomic levels following co-culture. The cells were also assessed at a physiological/functional level by evaluating their ability to form a functional intestinal epithelial barrier. This was achieved using an in vitro co-culture model with intestinal subepithelial myofibroblasts by measuring transepithelial resistance, permeability to protein and morphology in a simple tissue model co-culture. The cells did not display the morphological or physiological characteristics associated with intestinal epithelial cells in the model system. Overall this work has shown that co-culturing pluripotent mES cells with embryonic chick gut tissue can induce differentiation towards the ISC fate. Pre-treating the cells with growth factors in vitro did not seem to enhance this differentiation but there was scope to refine these techniques. Following the differentiation protocols the cells did not display the desired physiological characteristics but again there was scope to refine the techniques particularly with regard to selecting cells positive for the expression of the chosen molecular markers. These techniques show promise but do require some further development.

Peroxisome proliferators (PPs) cause proliferation of peroxisomes and hepatocarcinogenesis in rodent liver, mediated by Peroxisome Proliferator-Activated Receptor-alpha (PPARalpha). There are marked species differences in peroxisome proliferator-induced responses, and the functionality of PPARalpha may be an important determinant factor in species sensitivity to PPs. Primary hepatocytes were investigated for a highly responsive marker induced by PPs to study the effects of transfected PPARalphaCYP4A1 was highly induced in rat hepatocytes that require hydrocortisone for maximal induction. Hepatocytes were cultured in hydrocortisone-deficient media to determine if reduced endogenous PPARalpha was associated with lowered induction of CYP4A1. However, there was residual induction of CYP4A1 by peroxisome proliferators. Primary hepatocytes from PPARalpha knock-out (-/-) mice were investigated as they lack endogenous PPARalphaIn vitro and in vivo studies demonstrated that Cyp4a10 and 14 were highly inducible by PPs in the hepatocytes of wild-type but not in -/- mice. However, addition of either mouse or guinea pig PPARalpha in -/- hepatocytes did not induce the expression of these marker genes, although both receptors showed trans-activation ability in a reporter assay. The failure of added PPARalpha to activate endogenous genes responsive to PPs, whilst at the same time activating episomal DNA containing response elements of PP-inducible gene, suggests that the endogenous genes require PPARalpha to remain in an accessible conformation. Although hamster is considered to be a partially-responsive species to PPs, their response to PPs is poorly characterized. Three CYP4A genes (CYP4A17, 18 and 19) were cloned from hamster liver cDNA, and hepatic CYP4A17 was found to be highly inducible by PPs. In addition, PPARalphawas cloned from hamster liver and shows higher identity to rat and mouse PPARalpha than to human and guinea pig. Hepatic expression of PPARalpha mRNA was compared between mouse, hamster and guinea pig. The level of PPARalpha transcript was found to correlate well with species response to PPs, i.e. mouse (highly responsive species) has the highest level and guinea pig (non-responsive) the lowest, while hamster (partially-responsive) has an intermediate level. This is consistent with a model where the level of expression of hepatic PPARalpha determines species response to PPs. Expression of PPARalpha and transcriptional coactivators, such as PBP, SRC-1 and CBP/p300, were confined to mouse liver at the RNA level, but in each case expression showed homogenous distribution within the liver acinus and was non-inducible by PPs. Mouse PPARalpha ligand binding domain (LBD) was bacterially expressed as a histidine-tagged protein and soluble proteins were purified using affinity and column chromatography. Functional LBD may serve as a useful bait in protein-protein interaction studies for the identification of any novel PPARalpha interacting coactivator protein.

Philanthotoxin-433 (PhTX-433) is an active component of the Egyptian solitary digger wasp, Philanthus triangulum, venom which non-selectively inhibits several excitatory ion channels. With the aim of improving potency and selectivity, several synthetic analogues were developed based on the single or multiple modifications in the hydrophilic polyamine tail and hydrophobic aromatic head group of the PhTX. In the first part of this study, we investigated the pharmacological actions of two synthetic analogues, Philanthotoxin-343 (PhTX-343) and Philanthotoxin-12 (PhTX-12), on mammalian hetero- and homooligomeric nicotinic acetylcholine receptor (nAChR) subunit combinations expressed in Xenopus oocytes. Whole-cell currents in response to application of acetylcholine alone or co-applied with PhTX-analogue were studied electrophysiologically using two-electrode voltage-clamp at three different membrane holding potentials (VH = -60 mV, -80 mV and -100 mV). Concentration-inhibition curves were constructed and IC50 values estimated for each holding potential. The IC50 (95% CI, n=oocytes) values for PhTX-343 inhibition of α3β4, α3β2, α4β2, α4β4, α7, α3β4(F255V), α4β2(V253F) and α1β1δγ peak currents at -100 mV were 0.07 µМ (0.05-0.10 µМ, n=9), 3.20 µМ (1.69-6.26 µМ, n=8), 0.14 µМ (0.09-0.21 µМ, n=7), 0.28 µМ (0.17-0.46 µМ, n=6), 8.7 µМ (6.9-11.0 µМ, n =9), 0.01 µМ (0.01-0.02 µМ, n=8), 0.10 µМ (0.06-0.16 µМ, n=8) and 3.2 µМ (2.5-4.1 µМ, n=9) respectively; for PhTX-12 they were 2.03 µМ (1.22-3.26 µМ, n=8), 74.0 µМ (21-259 µМ, n=10), 1.6 µМ (0.7-3.6 µМ, n=11), 1.82 µМ (0.81-4.11 µМ, n=9), 12.1 µМ (9.0-16.0 µМ, n =10), 1.0 µМ (0.5-2.0 µМ, n=5), 3.4 µМ (1.3-8.7 µМ, n=9) and 5.0 µМ (2.8-8.8 µМ, n=8) respectively; i.e. in contrast to M-nAChR, PhTX-343 was more potent than PhTX-12 in all cases. All IC50s were lower when inhibition was measured after 1 minute of application indicating use dependence. For inhibition of heteromeric nAChRs, the peak potency of PhTX-343 was strongly augmented by holding the cell at more negative VH while this was not the case for PhTX-12 where only weak voltage-dependence was observed. The inhibition of homomeric α7 receptors by PhTX-343 was voltage-independent, whereas the block by PhTX-12 was voltage-dependent. In addition, these two synthetic analogues were different in their recovery rates (PhTX-12 was generally faster than PhTX-343) but they were similar in their mechanism of action, non-competitive inhibition. In the second part of the project, I explored the structure-activity relations of twenty one synthetic analogues, with a view to improving their activity and selectivity on rat neuronal α4β2 and α3β4 nAChRs at VH = -100 mV. We showed that the presence of positive charge in the polyamine tail of PhTX compounds is essential for nAChR subtype selectivity and its removal made the molecule lose its selectivity. Also, it appears that adding a bulky group to the terminal ammonium drastically reduced activity whereas a similar addition to the head region it increased their potency. In addition, we identified the key regions and substitutions responsible for increasing PhTX activity, cyclohexylalanine in place of tyrosine, and selectivity, phenyl group. Analogues having cyclohexylalanine and a phenolic group in the head region showed IC50 values in the low nano-molar and pico-molar (160-400 pM) range. These data suggest that PhTXs could serve as lead compounds for highly potent and selective inhibitors of N-nAChRs.

Self-renewal and pluripotency, the hallmarks of human embryonic stem cells (hESC), confer these cells with the capacity to expand indefinitely while maintaining the ability to differentiate into any cell type of the human body; thus, making hESC a valuable source of functional differentiated cells suitable for applications in regenerative medicine, drug discovery, biotechnology, biopharmaceuticals and developmental biology. However, the large-scale production of clinical-grade hESC, required for such applications, has been hampered by the current culture conditions in which hESC still depend on the use of mouse embryonic fibroblast-conditioned medium (MEF-CM) for their efficient growth. Therefore, investigation of the factors provided by MEFs is of the utmost importance to discover which components of MEF-CM allow the long-term expansion of undifferentiated hESC. While considerable progress has been made on the identification of the protein components of MEF-CM, very little is known about the small molecules (metabolites) secreted by MEFs. In this context, an untargeted metabolomics method was developed for the investigation of potential bioactive metabolites present in MEF-CM implicated in the proliferation and/or maintenance of pluripotency of hESC in vitro. A metabolomics method was applied and successfully identified a number of metabolites which were later confirmed in their identities with the use of authentic standards, to be further investigated for their effect on hESC culture. Interestingly, the addition of PGE2, 6-keto-PGF1α, 9, 12, 13-TriHOME, 7-Ketocholesterol and stearidonic acid (the metabolites found in MEF-CM) to the unconditioned medium (UM), a medium incapable of the maintenance of hESC, showed a delay in apoptosis when compared to the negative control UM; thus, suggesting that these metabolites could help with the proliferation of hESC. Increasing evidence that hESC secrete factors into their microenvironment that can also help them to proliferate or to maintain an undifferentiated state prompted the application of the same metabolomics method to the analysis of hESC spent culture media. The results identified lysophospholipids (LPLs) as potential molecules mediating some biological activities; however, the precise role of these LPLs still remains to be determined. Overall, the results of this thesis are expected to impact and add knowledge to the field of stem cell biology providing useful information for the creation and development of more efficient and defined culture conditions for the propagation of hESC with the appropriate quality to realise their widespread application in clinic and other research areas.

Malformations in humans at birth have been recorded since ancient times. These malformations are anatomical or physiological anomalies present at the time of birth that may be caused by genetic or environmental factors or a combination of both. The pathogenesis is only known in 10%, of which 1% or less are caused by drugs and medications. Certain disease states, like maternal epilepsy and depression during gestation itself, contribute to abnormal development. Further, this dilemma is augmented by the use of medications during pregnancy. The antiepileptic (AEDs) and antidepressant drugs (ADPs) with a history of producing malformed neonates are mostly classified as moderate teratogens. This study was designed to evaluate teratogenic potential at the cellular and molecular levels of AEDs and ADPs on cardiomyocytes at different stages of development and the neural stem cell derived neurons using in vitro systems. In the micromass system (MM), five day old embryonic chick cardiomyocytes were cultured to form beating foci, while embryonic stem cell were differentiated into contracting cardiomyocytes (ESDC) using the hanging drop method. In a third in vitro system early chick Neural Stem Cells (NSC) were diverted to a neuronal lineage. Drug toxic effects were estimated on cultured cell viability and protein content. The effects on gap junctions (Cx43) in cardiomyocytes and neurofilament (NF) in NSC were also evaluated because of their important role in cell differentiation and regulation. Oxidative stress, being the potential source of xenobiotic toxicity induction, was also analysed and toxic effects were counteracted using antioxidants and other molecules. In AEDs, valproic acid (VPA) mainly targeted the cardiomyocyte differentiation and contractile activity with reduced Cx43 turnover. In NSC the VPA effects were different and it did not inhibit the neuronal differentiation. With carbamazepine (CBZ) the low doses showed no effect on NSC compared to high doses. In ESDC, the contractile activity stops at a 200µM dose with reduced cell viability and proliferation. Cx43 phosphorylation was reduced after CBZ treatment which might have affected the contractile activity. An increase ROS production with CBZ treatment was recorded, which was protected either by the addition of Ascorbic acid (AA) or superoxide dismutase (SOD). The other AEDs, Phenytoin (PHT) and Primidone (PRM), mainly affected the cardiomyocyte contractile activity with some chronic exposure effects. In ADP, bupropion (BPN) severely affects cell proliferation in all systems. The NF-L was not statistically reduced in neurons but Cx43 expression in cardiomyocytes declined which might result in reduced contraction. The other ADP, lithium carbonate showed developmental stage dependent effect on cardiogenesis, where contractile activity ceased completely at higher dose in the ESDC with increased cell proliferation. Lithium mimics the Wnt/β-catenin pathway and also inhibits the PI cycle, effects which were reversed by the addition of myo-inositol in the ESDC system. In NSC the lithium showed no significant inhibitory effects on neural differentiation at and above drug serum therapeutic concentrations. The active constituents of the herbal antidepressant drug St. John’s wort, hypericin and hyperforin, showed synergistic inhibition of contractile activity with reduced proliferation at higher doses in the MM system. Drug interference at the molecular level during development may induce modification at the gene and protein levels with altered signalling. The tissue specific effects depend on the drug mechanism, while increased oxidative stress in part has a contribution in initiating the embryopathies. By identifying the exact mechanism of toxicity induction, the molecular mechanism can be protected against and thus abnormal development be avoided.

Of the numerous l3-blockers clinically available to treat conditions such as angina pectoris, hypertension and heart failure, none possess antagonist activity specific to the beta1-adrenoceptor. Those described as 'cardioselective', such as nebivolol and bisoprolol, generally show less than 50-fold beta1/beta2-selectivity, which can be lost at higher doses. Others, such as propranolol and sotalol are actually more beta2-selective. Overall, a degree of concomitant beta2-adrenoceptor blockade (risking compromised respiratory function and loss of peripheral vasodilatation) by current therapeutic agents precludes their use in patients with disorders such as asthma and peripheral vascular disease. This project aims to develop novel molecules with much improved beta1/beta2-selectivity over current beta1-blocker therapy as well as improving knowledge of ligand-receptor interaction at the beta1-adrenoceptor, through an analogue-based drug discovery approach. A highly selective or specific beta1-adrenoceptor antagonist is likely to cause fewer side-effects and be suitable for use in previously contraindicated disease states. This thesis reports the design, synthesis and pharmacological data (provided by Dr. Jillian Baker) of a library of novel ligands for the beta1- adrenoceptor, based upon the lead compound LK 204-545. LK 204-545 was selected based on reported high potency at the beta1-adrenoceptor as well as good beta1/beta2-selectivity. Modification of various motifs on structures derived from LK 204-545 allowed the generation of new structure-activity relationships and ultimately afforded the highly 131-adrenoceptor selective compound, 1-(2-(3-(4-(2-(cyclopropylmethoxy)ethoxy)phenoxy)-2-hydroxypropyl amino)ethyl)-3-(4-hydroxyphenyl)urea hydroformate (12Sc). This compound acted as a highly-selective beta1-adrenoceptor antagonist in a pilot in-vivo study in the regional hemodynamic rat model (carried out by Prof. Sheila Gardiner).

Characterisation of expression of functional Peroxisome Proliferator Activated Receptora (PPARalpha)receptor in rodent species responsive and non-responsive to peroxisome proliferators is important for our understanding of the molecular mechanism of peroxisome proliferation and peroxisome proliferator induced hepatocarcinogenesis. In vitro electromobility shift assays, demonstrated that rodent liver nuclear proteins (LNP) bound to a Peroxisome Proliferator Response Element (PPRE) in a sequence specific manner and that LNP from methylclofenapate (MCP) treated mice do not have enhanced binding to a PPRE. These results demonstrate that in MCP treated mice, PPAR alpha levels with functional DNA binding do not increase. The diurnal expression of mouse PPAR alpha (mPPARalpha) protein in liver was examined by western blotting. There was no observable difference in the expression of mPPARalpha across a 24 hour period. In C57 BL/6 mice, PPARalpha protein levels are not regulated in a diurnal manner. A comparison of mouse and guinea pig LNP revealed a PPARalpha-immunoreactive protein in guinea pig. Guinea Pig PPARalpha (gPPAR a) was cloned and found to encode a 467 amino acid protein. Phylogenetic analysis of gPPARalpha showed a high substition rate: maximum likelihood analysis was consistent with rodent monophyly, but could not exclude rodent polyphyly (p~0.07). The gPPAR alpha cDNA was expressed in 293 cells, and mediated the induction of the luciferase reporter gene by the peroxisome proliferator Wy-14,643, dependent upon the presence of a PPRE. The gPPAR alpha mRNA and protein was expressed in guinea pig liver, although at lower levels compared to PPAR alpha expression in mice. The evidence presented here supports the idea that guinea pigs serve as a useful model for human responses to peroxisome proliferators. mPPAR alpha DNA binding domain (mPPARalpha-DBD) was cloned and expressed as a fusion protein. Both His*6-mPPARalpha-DBD and thioredoxin-mPPARalpha-DBD were produced as insoluble proteins when over expressed in E.coli. In vitro translated mPPAR alpha-DBD did not bind to a PPRE in an electromobility shift assay.

The solar UVB radiation incident on a horizontal surface was measured and related to more routinely recorded meteorological variabIes in a study of the UVB climatology of the English East Midlands. Exposure of individuals in this climate was monitored and related to vitamin D status. On clear days relations were found between the logarithm of UVB intensity lƛ and airmass µ,and at 304 nm where ozone amount [O3] is the dominant atmospheric attenuating factor a2 In Iƛ/aµa[03] was close to the ozone absorption coefficient for this wavelength. At longer wavelengths other attenuation processes have to be accounted for. Measurements of the waveband 300-316 nm were compared with irradiation over broader wavebands. On clear days the ratio of UVB to visible irradiance IB/IV was 4.16 cos z + x where z is the solar zenith angle and x is a coefficient which varies from day to day. Similar analysis for the full solar waveband IF showed a similar linearity of IB/IF with cos z for each day, but both slope and intercept changed between days. A relation between daily integrated totals of UVB and full solar radiation (300-3000 nm) was found, enabling UVB radiation to be estimated from measurements made with a standard meteorological pyranometer. The best estimates require daily figures for ozone concentration but an approximation may still be made using monthly mean concentrations or climatological averages. Diffuse UVB radiation was measured and found to be always greater than 0.5 global UVB. The shade-ring correction applicable in this region of the spectrum is ~ 0.01 greater than the geometric correction. Estimates of the anisotropy of UVB sky radiation gave the relative strength of the circumsolar region as 0.36 with an angular width of 0.78 radians. Polysulphone film was tested and found suitable for use as a personal dosimeter for solar UVB radiation. The UVB exposure of elderly long-stay hospital patients was monitored for a three month period and compared with that of a young healthy population. Plasma 25-hydroxyvitamin D concentrations were measured to assess vitamin D status and the change in plasma 25(OH)D resulting from skin irradiated with solar UVB was found to be 6.9 ± 0.4 ng J -1 for the elderly and + 7.3 ± 3.4 ng J-1 for the young volunteers suggesting little difference between the responses of the elderly and the young. The implication of these figures is that sunlight exposure: of a few hours per week is adequate to maintain a healthy vitamin D status.

DEF6 is a Rho-Guanine Nucleotide Exchange Factor (GEF) with a poorly characterised role in T-cell receptor signalling and unknown structure1-3. Its importance in T-cell signalling has been highlighted by observations that DEF6-deficient T cells display defective actin polymerisation and polarisation, have a decreased susceptibility to CD3-induced apoptosis4, and exhibit aberrant expression of the inflammatory cytokine IL-172. DEF6 is a GEF for Rho-family GTPases including CDC42, RhoA and Rac11, which unlike other Rho-GEFs exhibits a reversal of the N-C terminal orientation of the DH and PH domains only found in DEF6 and its homolog in B-cells, SWAP70. Through this role DEF6 has been shown to be involved in the spatiotemporal organisation of T cell signalling through the GTPase CDC425. DEF6-deficient mice have been shown to develop autoimmune diseases and have also been shown to be resistant to the development of other autoimmune diseases, indicating a loss of control of the immune response. The molecular mechanism behind these phenotypes remains to be identified. Exogenously expressed DEF6 has been shown to form cytoplasmic granules under conditions of arrested translation, similar to processing bodies and stress granules which are involved in translational regulation. Data presented here furthers this work and demonstrates endogenous DEF6 forms granules in activated T-cells. Exogenously expressed DEF6 shows partial co-localisation with markers of P-bodies, and demonstrates slow recovery from FRAP indicating a structural role within granules. The granules move within the cell, and analysis of this movement suggests the granules move via active transport, suggesting a link to the microtubule network. Examination of the individual domains of DEF6 demonstrates that the N-terminus is likely to have a role in the co-localisation of DEF6 with P-bodies, whilst the C-terminal DHL domain drives aggregation of DEF6 into granules. Data here confirms the predicted formation of a coiled-coil structure within the DHL domain.

Cell-based metabolomics using LC-MS systemizes the study of the uniqueness of small-molecule metabolite (metabolomes) profiles in cellular processes. Cell-based metabolomics can potentially be used in many applications for the study of biological perturbation from stimulants in cellular pathways. The advantages of cell-based metabolomics include ease of control and interpretation when compared to the study of human subjects and animal models. Furthermore, this method can decrease some highly challenging problems that occur in genomics, transcriptomics and proteomics. Nowadays, cell culture in metabolomics studies has been used in many applications. These include cell culture and bioreactor optimisation, phenotype classification, stimulant testing effect, target and toxicity analysis, metabolic networks determination and modelling, and biomarker and drug target discovery. In this study, the reverse phase-liquid chromatography-mass spectrometry and hydrophilic interaction chromatography-mass spectrometry for comprehensive metabolic profiling well suited to the untargeted analysis of non-polar and polar metabolites in mammalian cells were developed, optimized and validated. These methods can separate and detect most of hydrophobic and polar metabolites that are normally found in mammalian cell lines. After that the LC-MS methods were applied to assess the effects of drugs with known and unknown cellular metabolic effects on three mammalian cell lines, namely HMVECs for antipsychotics experiment, MCF-7 cells for cordycepin experiment and MIN6 cells for fluoxetine experiment by using untargeted metabolic profiling. The global effects of antipsychotics at high therapeutic dosage in HMVECs were investigated. The results support for the toxicity hypothesis with measurements that confirm previous findings and reveal the exact biological pathways of antipsychotic-altered BBB functions. It was found that antipsychotics may affect the bioenergetics pathway due to mitochondrial dysfunction resulted in ketoacidosis and inducing oxide stress by reactive oxygen species generation. In the MCF- cell experiment, the results of the untargeted metabolite profiling demonstrated the clear anti-breast cancer effects of cordycepin and pentostatin. By investigating the metabolite profiles, clear synergistic effects of cordycepin and pentostatin combined in comparison to cordycepin activity alone in MCF-7 cells was observed. Furthermore, the pathway analysis indicated that anti-breast cancer activity was mainly responsible for alterations in purine and pyrimidine metabolism and bioenergetics. Additionally, cordycepin may be involved in the inhibition of cell proliferation and differentiation, and the activation of cell apoptosis. The last experiment on MIN6 cells, the developed and optimized HILIC-MS approach in order to determine the biological pathways which are impaired by fluoxetine on glucose-stimulated insulin secretion on MIN6 cell lines was performed. It is found that fluoxetine may impair glycolysis, TCA and fatty acid metabolism on MIN6 cell lines. Moreover, it is also reveal that the alteration of biological pathways on MIN6 cells by known ETC inhibitors (rotenone (Complex I inhibitor) antimycin (Complex III inhibitor)) and azide (a complex IV inhibitor). From comparison with these ETC inhibitors, it is found that fluoxetine may have the same effect pattern with azide.