A comparison of Markov and generalized linear models of hospital mortality

This paper reports a comparison of three modeling strategies for the analysis of hospital mortality in a sample of general medicine inpatients in a Department of Veterans Affairs medical center. Logistic regression, a Markov chain model, and longitudinal logistic regression were evaluated on predictive performance as measured by the c-index and on accuracy of expected numbers of deaths compared to observed. The logistic regression used patient information collected at admission; the Markov model was comprised of two absorbing states for discharge and death and three transient states reflecting increasing severity of illness as measured by laboratory data collected during the hospital stay; longitudinal regression employed Generalized Estimating Equations (GEE) to model covariance structure for the repeated binary outcome. Results showed that the logistic regression predicted hospital mortality as well as the alternative methods but was limited in scope of application. The Markov chain provides insights into how day to day changes of illness severity lead to discharge or death. The longitudinal logistic regression showed that increasing illness trajectory is associated with hospital mortality. The conclusion is reached that for standard applications in modeling hospital mortality, logistic regression is adequate, but for new challenges facing health services research today, alternative methods are equally predictive, practical, and can provide new insights. ^

FUNCTIONS OF DEADENYLATION FACTORS IN MRNA DECAY AND MRNA PROCESSING BODY FORMATION

Most newly synthesized messenger RNAs possess a 5’ cap and a 3’ poly(A) tail. The process of poly(A) tail shortening, also termed deadenylation, is important for post-transcriptional gene regulation, because deadenylation not only leads to mRNA translational inhibition but also is the first step of major mRNA degradation. Translationally inhibited mRNAs can be stored and/or degraded in dynamic cytoplasmic foci termed mRNA processing bodies, or P bodies, which are conserved in eukaryotes. To shed new light on the mechanisms of P body formation and P body functions, I focused on the link between deadenylation factors and P bodies. I found that the two major deadenylation complexes, Pan3-Pan2 and Ccr4-Caf1, can both be enriched in P bodies. The deadenylase activity of the Ccr4-Caf1 complex is prerequisite for P body formation. Pan3, but not the deadenylase Pan2, is essential for P body formation. While the C-terminal domain of Pan3 is important for interaction with Pan2, Pan3 N-terminal domain is important for Pan3 to form cytoplasmic foci colocalizing with P bodies and to promote mRNA decay. Interestingly, Pan3 N-terminal domain may be phosphorylated to regulate Pan3 localization and functions. Aside from the functions of the two deadenylation complexes in P bodies, I also studied all reported human P body proteins as a whole using bioinformatics. This effort not only has generated a comprehensive picture of the functions of and interactions among human P body proteins, but also has predicted proteins that may regulate P body formation and/or functions. In summary, my study has established a direct link between mRNA deadenylation and P body formation and has also led to new hypotheses to guide future research on how P body dynamics are controlled.

NEW TARGET GENES FOR TUMOR SUPPRESSORS p53 AND p73 IN REGENERATING LIVER

The p53-family of proteins regulates expression of target genes during tissue development and differentiation. Within the p53-family, p53 and p73 have hepatic-specific functions in development and tumor suppression. Despite a growing list of p53/p73 target genes, very few of these have been studied in vivo, and the knowledge regarding functions of p53 and p73 in normal tissues remains limited. p53+/-p73+/- mice develop hepatocellular carcinoma (HCC), whereas overexpression of p53 in human HCC leads to tumor regression. However, the mechanism of p53/p73 function in liver remains poorly characterized. Here, the model of mouse liver regeneration is used to identify new target genes for p53/p73 in normal quiescent vs. proliferating cells. In response to surgical removal of ~2/3 of liver mass (partial hepatectomy, PH), the remaining hepatocytes exit G0 of cell cycle and undergo proliferation to reestablish liver mass. The hypothesis tested in this work is that p53/p73 functions in cell cycle arrest, apoptosis and senescence are repressed during liver regeneration, and reactivated at the end of the regenerative response. Chromatin immunoprecipitation (ChIP), with a p73-antibody, was used to probe arrayed genomic sequences (ChIP-chip) and uncover 158 potential targets of p73-regulation in normal liver. Global microarray analysis of mRNA levels, at T=0-48h following PH, revealed sets of genes that change expression during regeneration. Eighteen p73-bound genes changed expression after PH. Four of these genes, Foxo3, Jak1, Pea15, and Tuba1 have p53 response elements (p53REs), identified in silico within the upstream regulatory region. Forkhead transcription factor Foxo3 is the most responsive gene among transcription factors with altered expression during regenerative, cellular proliferation. p53 and p73 bind a Foxo3 p53RE and maintain active expression in quiescent liver. During liver regeneration, binding of p53 and p73, recruitment of acetyltransferase p300, and an active chromatin structure of Foxo3 are disrupted, alongside loss of Foxo3 expression. These parameters of Foxo3 regulation are reestablished at completion of liver growth and regeneration, supporting a temporary suspension of p53 and p73 regulatory functions in normal cells during tissue regeneration.

Analysis of Band 4.1B in Integrin-Mediated Cell Adhesion and Signaling

Band 4.1B is a cytoskeletal adaptor protein that regulates various cellular behavior; however, the mechanisms by which Band 4.1B contributes to intracellular signaling are unclear. This project addresses in vivo and in vitro functions for Band 4.1B in integrin-mediated cell adhesion and signaling. Band 4.1B has been shown to bind to β8 integrin, although cooperative functions of these two proteins have not been determined. Here, functional links between β8 integrin and Band 4.1B were investigated using gene knockout strategies. Ablation of β8 integrin and Band 4.1B genes resulted in impaired cardiac morphogenesis, leading to embryonic lethality by E11.5. These embryos displayed malformation of the outflow tract that was likely linked to abnormal regulation of cardiac neural crest migration. These data indicate the importance of cooperative signaling between β8 integrin and Band 4.1B in cardiac development. The involvement of Band 4.1B in integrin-mediated cell adhesion and signaling was further demonstrated by studying its functional roles in vitro. Band 4.1B is highly expressed in the brain, but its signaling in astrocytes is not understood. Here, Band 4.1B was shown to promote cell spreading likely by interacting with β1 integrin via its band 4.1, ezrin, radixin, and moesin (FERM) domain in cell adhesions. In astrocytes, both Band 4.1B and β1 integrin were expressed in cell-ECM contact sites during early cell spreading. Exogenous expression of Band 4.1B, especially its FERM domain, enhanced cell spreading on fibronectin, an ECM ligand for β1 integrin. However, the increased cell spreading was prohibited by blocking β1 integrin. These findings suggest that Band 4.1B is crucial for early adhesion assembly and/or signaling that are mediated by β1 integrin. Collectively, this study was the first to establish Band 4.1B as a modulator of integrin-mediated adhesion and signaling.

MOLECULAR MECHANISMS UNDERLYING THE TRANSCRIPTIONAL REGULATION OF T HELPER 17 AND REGULATORY T CELLS

CD4+ T helper (Th) lymphocytes are vital for integrating immune responses by orchestrating the function of other immune cell types. Naïve Th cells can differentiate into different effector subsets that are characterized by their cytokine profile and immune regulatory functions. These subsets include Th1, Th2, Th17, natural and inducible regulatory T cells (nTreg and iTreg respectively), among others. We focused our investigation on two Th lineages, Th17 and regulatory T cells, with opposing functions in the immune system. These subsets have been suggested to be reciprocally regulated since they both require TGF-b for their development. We investigated the role of the Treg-associated master transcription factor Foxp3, and found that Foxp3 inhibits Th17 cell generation by preventing the transcriptional activity of the two main Th17-specific transcription factors, nuclear orphan receptors RORa and RORgt. At the molecular level, we identified two different functional domains in Foxp3 required for such inhibition: the LQALL sequence in exon 2 and the TIP60/HDAC7 binding domain. These domains could be crucial to either prevent the association of the nuclear receptors to coactivators or to recruit histone deacetylases to RORa- or RORgt-target genes. Since TGF-b is a common cytokine required for the commitment towards both Th lineages, we determined the role of the TGF-b-dependent signaling pathway in the generation of each subset. By using mice with deficiencies in signaling molecules downstream of TGF-b, we found that while Smad2, Smad3 and Smad4 are required for the generation of iTreg cells, only Smad2 is indispensable for the induction of IL-17-producing cells, suggesting that TGF-b induces these T helper lineages through differential signaling pathways. Thus, our findings describe novel transcriptional regulatory mechanisms that control the generation of two T helper lineages with opposing functions. These findings could provide novel therapeutic targets to treat diseases where the balance of these T cells is dysregulated, such as in autoimmunity, chronic infectious diseases and cancer.

Expression, induction and regulation of the cytochrome P450 monooxygenase system in the rat glioma C6 cell line

The cytochrome P450 monooxygenase system consists of NADPH- cytochrome P450 reductase (P450 reductase) and cytochromes P450, which can catalyze the oxidation of a wide variety of endogenous and exogenous compounds, including steroid hormones, fatty acids, drugs, and pollutants. The functions of this system are as diverse as the substrates. P450 reductase transfers reducing equivalents from NADPH to P450, which in turn catalyzes metabolic reactions. This enzyme system has the highest level of activity in the liver. It is also present in other tissues, including brain. The functions of this enzyme system in brain seem to include: neurotransmission, neuroendocrinology, developmental and behavioral modulation, regulation of intracellular levels of cholesterol, and potential neurotoxicity.^ In this study, we have set up the rat glioma C6 cell line as an in vitro model system to examine the expression, induction, and tissue-specific regulation of P450s and P450 reductase. Rat glioma C6 cells were treated with P450 inducers phenobarbital (PB) or benzo(a)anthracene (BA). The presence of P450 reductase and of cytochrome P450 1A1, 1A2, 2A1, 2B1/2, 2C7, 2D1-5 and 2E1 was detected by reverse transcription followed by polymerase chain reaction (RT-PCR) and confirmed by restriction digestion. The induction of P450 1A1 and 2B1/2 and P450 reductase was quantified using competitive PCR. Ten- and five-fold inductions of P450 1A and 2B mRNA after BA or PB treatments, respectively, were detected. Western blot analysis of microsomal preparations of glioma C6 cells demonstrated the presence of P450 1A, 2B and P450 reductase at the protein level. ELISAs showed that BA and PB induce P450 1A and 2B proteins 7.3- and 13.5-fold, respectively. Microsomes prepared from rat glioma C6 cells showed cytochrome P450 CO difference spectra with absorption at or near 450 nm. Microsomes prepared from rat glioma C6 cells demonstrated much higher levels of ethoxyresorufin O-deethylase (EROD) and pentoxyresorufin O-dealkylase (PROD) activity, when treated with BA or PB, respectively. These experiments provide further evidence that the rat glioma C6 cell line contains an active cytochrome P450 monooxygenase system which can be induced by P450 inducers. The mRNAs of P450 1A1 and 2B1/2 can not bind to the oligo(dT) column efficiently, indicating they have very short poly(A) tails. This finding leads us to study the tissue specific regulation of P450s at post-transcriptional level. The half lives of P450 1A1 and 2B1/2 mRNA in glioma C6 cells are only 1/10 and 1/3 of that in liver. This may partly contribute to the low expression level of P450s in glial cells. The induction of P450s by BA or PB did not change their mRNA half lives, indicating the induction may be due to transcriptional regulation. In summary of this study, we believe the presence of the cytochrome P450 monooxygenase system in glial cells of the brain may be important in chemotherapy and carcinogenesis of brain tumors. ^

The role of functional tumor suppressor 15-lipoxygenase 2 (15-LOX2) in normal human prostate epithelial cells and prostate cancer development

15-Lipoxygenase 2 (15-LOX2) is a recently cloned human lipoxygenase that shows tissue-restricted expression in prostate, lung, skin, and cornea. The protein level and enzymatic activity of 15-LOX2 have been shown to be down-regulated in prostate cancers compared with normal and benign prostate tissues. We report the cloning and functional characterization of 15-LOX2 and its three splice variants (termed 15-LOX2sv-a, 15-LOX2sv-b, and 15-LOX2sv-c) from primary prostate epithelial (NHP) cells. Western blotting with multiple NHP cell strains and prostate cancer (PCa) cell lines reveals that the expression of 15-LOX2 is lost in all PCa cell lines, accompanied by decreased enzymatic activity. 15-LOX2 is expressed at multiple subcellular locations, including cytoplasm, cytoskeleton, cell-cell border, and nucleus. Surprisingly, the three splice variants of 15-LOX2 are mostly excluded from the nucleus. To elucidate the relationship between nuclear localization, enzymatic activity, and tumor suppressive functions, we established PCa cell clones stably expressing 15-LOX2 or 15-LOX2sv-b. The 15-LOX2 clones express 15-LOX2 in the nuclei and possess robust enzymatic activity, whereas 15-LOX2sv-b clones show neither nuclear protein localization nor arachidonic acid-metabolizing activity. Interestingly, both 15-LOX2- and 15-LOX2sv-b-stable clones proliferate much slower in vitro when compared with control clones. When orthotopically implanted in nude mouse prostate, both 15-LOX2 and 15-LOX2sv-b suppress PC3 tumor growth in vivo. Finally, cultured NHP cells lose the expression of putative stem/progenitor cell markers, slow down in proliferation, and enter senescence. Several pieces of evidence implicate 15-LOX2 plays a role in replicative senescence of NHP cells: (1) promoter activity and the mRNA and protein levels of 15-LOX2 and its splice variants are upregulated in serially passaged NHP cells, which precede replicative senescence and occur in a cell-autonomous manner; (2) PCa cells stably expressing 15-LOX2 or 15-LOX2sv-b show a passage-related senescence-like phenotype; (3) enforced expression of 15-LOX2 or 15-LOX2sv-b in young NHP cells induce partial cell-cycle arrest and senescence-like phenotypes. Together, these results suggest that 15-LOX2 suppress prostate tumor development and do not necessarily depend on arachidonic acid-metabolizing activity and nuclear localization. Also, 15-LOX2 may serve as an endogenous prostate senescence gene and its tumor-suppressing functions might be associated with its ability to induce cell senescence. ^

Identification and characterization of new genes involved in Drosophila heart development

Heart development is a crucial and conserved process that is related to the major type of human birth defects. Dorsal vessel, the Drosophila heart, has been regarded as an insightful system to identify new genes and study gene functions involved in heart development. Using heart-specific GFP transgenes, I did a genetic screen for cardiogenic genes on Drosophila chromosome II. Drosophila mutants that carry chromosome II deficiencies were tested for their phenotypes of heart development. Based on the screen results, chromosome regions containing genes required for heart development were identified. Fly strains with single gene mutations located within the defined deficiency regions were tested further. Seven genes have been identified to be involved in heart development. ^ The LIM homeodomain transcription factor gene tailup (tup) was further studied for its function in heart development. Based on this study, tup is expressed in cardioblasts and pericardial cells of the heart tube, as well as in associated lymph glands and alary muscles. In depth analysis of tup mutant phenotypes demonstrated tup is required for normal development of both heart and lymph glands. Tup was shown to bind to two DNA recognition sequences in the dorsal vessel enhancer of the Hand bHLH transcription factor gene, with one site proven essential for the expression of Hand in lymph glands, pericardial cells, and Svp/Doc cardioblasts. Together, these studies demonstrate that Tup is a critical new transcription factor in dorsal vessel morphogenesis and lymph gland formation, and strongly suggest Tup is a direct regulator of the expression of Hand in these developmental processes. ^

Analysis of life expectancy using linear and nonlinear models

Life expectancy has consistently increased over the last 150 years due to improvements in nutrition, medicine, and public health. Several studies found that in many developed countries, life expectancy continued to rise following a nearly linear trend, which was contrary to a common belief that the rate of improvement in life expectancy would decelerate and was fit with an S-shaped curve. Using samples of countries that exhibited a wide range of economic development levels, we explored the change in life expectancy over time by employing both nonlinear and linear models. We then observed if there were any significant differences in estimates between linear models, assuming an auto-correlated error structure. When data did not have a sigmoidal shape, nonlinear growth models sometimes failed to provide meaningful parameter estimates. The existence of an inflection point and asymptotes in the growth models made them inflexible with life expectancy data. In linear models, there was no significant difference in the life expectancy growth rate and future estimates between ordinary least squares (OLS) and generalized least squares (GLS). However, the generalized least squares model was more robust because the data involved time-series variables and residuals were positively correlated. ^

Effects of overweight and obesity on economic outcomes among U.S. adults with kidney disease

Background. Kidney disease is a growing public health phenomenon in the U.S. and in the world. Downstream interventions, dialysis and renal transplants covered by Medicare's renal disease entitlement policy in those who are 65 years and over have been expensive treatments that have been not foolproof. The shortage of kidney donors in the U.S. has grown in the last two decades. Therefore study of upstream events in kidney disease development and progression is justified to prevent the rising prevalence of kidney disease. Previous studies have documented the biological route by which obesity can progress and accelerate kidney disease, but health services literature on quantifying the effects of overweight and obesity on economic outcomes in the context of renal disease were lacking. Objectives . The specific aims of this study were (1) to determine the likelihood of overweight and obesity in renal disease and in three specific adult renal disease sub-populations, hypertensive, diabetic and both hypertensive and diabetic (2) to determine the incremental health service use and spending in overweight and obese renal disease populations and (3) to determine who financed the cost of healthcare for renal disease in overweight and obese adult populations less than 65 years of age. Methods. This study was a retrospective cross-sectional study of renal disease cases pooled for years 2002 to 2009 from the Medical Expenditure Panel Survey. The likelihood of overweight and obesity was estimated using chi-square test. Negative binomial regression and generalized gamma model with log link were used to estimate healthcare utilization and healthcare expenditures for six health event categories. Payments by self/family, public and private insurance were described for overweight and obese kidney disease sub-populations. Results. The likelihood of overweight and obesity was 0.29 and 0.46 among renal disease and obesity was common in hypertensive and diabetic renal disease population. Among obese renal disease population, negative binomial regression estimates of healthcare utilization per person per year as compared to normal weight renal disease persons were significant for office-based provider visits and agency home health visits respectively (p=0.001; p=0.005). Among overweight kidney disease population health service use was significant for inpatient hospital discharges (p=0.027). Over years 2002 to 2009, overweight and obese renal disease sub-populations had 53% and 63% higher inpatient facility and doctor expenditures as compared to normal weight renal disease population and these result were statistically significant (p=0.007; p=0.026). Overweigh renal disease population had significant total expenses per person per year for office-based and outpatient associated care. Overweight and obese renal disease persons paid less from out-of-pocket overall compared to normal weight renal disease population. Medicare and Medicaid had the highest mean annual payments for obese renal disease persons, while mean annual payments per year were highest for private insurance among normal weight renal disease population. Conclusion. Overweight and obesity were common in those with acute and chronic kidney disease and resulted in higher healthcare spending and increased utilization of office-based providers, hospital inpatient department and agency home healthcare. Healthcare for overweight and obese renal disease persons younger than 65 years of age was financed more by private and public insurance and less by out of pocket payments. With the increasing epidemic of obesity in the U.S. and the aging of the baby boomer population, the findings of the present study have implications for public health and for greater dissemination of healthcare resources to prevent, manage and delay the onset of overweight and obesity that can progress and accelerate the course of the kidney disease.^

Evaluation of PRESAGE® dosimeters for brachytherapy sources and the 3D dosimetry and characterization of the new AgX100 125I seed model

With continuous new improvements in brachytherapy source designs and techniques, method of 3D dosimetry for treatment dose verifications would better ensure accurate patient radiotherapy treatment. This study was aimed to first evaluate the 3D dose distributions of the low-dose rate (LDR) Amersham 6711 OncoseedTM using PRESAGE® dosimeters to establish PRESAGE® as a suitable brachytherapy dosimeter. The new AgX100 125I seed model (Theragenics Corporation) was then characterized using PRESAGE® following the TG-43 protocol. PRESAGE® dosimeters are solid, polyurethane-based, 3D dosimeters doped with radiochromic leuco dyes that produce a linear optical density response to radiation dose. For this project, the radiochromic response in PRESAGE® was captured using optical-CT scanning (632 nm) and the final 3D dose matrix was reconstructed using the MATLAB software. An Amersham 6711 seed with an air-kerma strength of approximately 9 U was used to irradiate two dosimeters to 2 Gy and 11 Gy at 1 cm to evaluate dose rates in the r=1 cm to r=5 cm region. The dosimetry parameters were compared to the values published in the updated AAPM Report No. 51 (TG-43U1). An AgX100 seed with an air-kerma strength of about 6 U was used to irradiate two dosimeters to 3.6 Gy and 12.5 Gy at 1 cm. The dosimetry parameters for the AgX100 were compared to the values measured from previous Monte-Carlo and experimental studies. In general, the measured dose rate constant, anisotropy function, and radial dose function for the Amersham 6711 showed agreements better than 5% compared to consensus values in the r=1 to r=3 cm region. The dose rates and radial dose functions measured for the AgX100 agreed with the MCNPX and TLD-measured values within 3% in the r=1 to r=3 cm region. The measured anisotropy function in PRESAGE® showed relative differences of up to 9% with the MCNPX calculated values. It was determined that post-irradiation optical density change over several days was non-linear in different dose regions, and therefore the dose values in the r=4 to r=5 cm regions had higher uncertainty due to this effect. This study demonstrated that within the radial distance of 3 cm, brachytherapy dosimetry in PRESAGE® can be accurate within 5% as long as irradiation times are within 48 hours.

The structural and functional relationship of the p53 tumor suppressor protein

The tumor-suppressing function of p53 can be affected in a variety of manners. Here, we describe a novel mechanism of transformation by mutant p53. Previously, it had been believed that mutant p53 molecules transform cells by oligomerizing with wild-type p53 and inactivating it. However, we demonstrated that there exists an additional mechanism of inactivation of p53 available to p53 mutants. It involves sequestration of cofactors necessary to p53, and subsequent interruption of its transactivation and tumor suppression functions. The p53 amino or carboxyl termini, known to interact with a large number of cellular factors, can affect wild-type p53 in this manner. Although they are unable to oligomerize with wild-type p53, they transform cells containing p53, and inhibit its transactivation ability. In addition, they interrupt growth suppression by p53, but not RB, confirming that they specifically affect p53 function, rather than having a general growth-stimulatory phenomenon. Also, we have cloned a p53 tumor mutation which results in expression of the amino terminus of p53. This provides a means to study the factor-sequestration transforming mechanism in vivo. Additionally, we found that the published sequence of the mdm2 gene is in error. mdm2 is a gene intimately involved with p53, blocking its ability to transform cells. Finally, previous data had established the influence of cell-cycle status on p53 function. In growth-arrested cells, wild-type p53 expressed by a transgene cannot activate transcription, but if these cells are forced to cycle by addition of cyclin E, p53 once again becomes functional. In this study, we extend these findings by examining only those cells successfully transfected, using fluorescence-activated cell sorting. Our results support the previous data, that cyclin E pushes growth-arrested cells back into the cell cycle. In summary, we have demonstrated the potential importance of cofactor association and protein modification to the abilities of p53 to cause transcription activation and repression, inhibition of DNA replication and induction of DNA repair, and initiation of cell-cycle arrest and apoptosis. Further elucidation of these processes and their roles in tumor suppression will prove fascinating indeed. ^