Improved cerebellar tissue classification on magnetic resonance images of brain.

PURPOSE: To develop and implement a method for improved cerebellar tissue classification on the MRI of brain by automatically isolating the cerebellum prior to segmentation. MATERIALS AND METHODS: Dual fast spin echo (FSE) and fluid attenuation inversion recovery (FLAIR) images were acquired on 18 normal volunteers on a 3 T Philips scanner. The cerebellum was isolated from the rest of the brain using a symmetric inverse consistent nonlinear registration of individual brain with the parcellated template. The cerebellum was then separated by masking the anatomical image with individual FLAIR images. Tissues in both the cerebellum and rest of the brain were separately classified using hidden Markov random field (HMRF), a parametric method, and then combined to obtain tissue classification of the whole brain. The proposed method for tissue classification on real MR brain images was evaluated subjectively by two experts. The segmentation results on Brainweb images with varying noise and intensity nonuniformity levels were quantitatively compared with the ground truth by computing the Dice similarity indices. RESULTS: The proposed method significantly improved the cerebellar tissue classification on all normal volunteers included in this study without compromising the classification in remaining part of the brain. The average similarity indices for gray matter (GM) and white matter (WM) in the cerebellum are 89.81 (+/-2.34) and 93.04 (+/-2.41), demonstrating excellent performance of the proposed methodology. CONCLUSION: The proposed method significantly improved tissue classification in the cerebellum. The GM was overestimated when segmentation was performed on the whole brain as a single object.

Multisite phosphorylation of ornithine decarboxylase in transformed macrophages

Ornithine decarboxylase (ODC), the initial inducible enzyme in the polyamine biosynthetic pathway, exists in the transformed macrophage RAW264 cell line as a phosphoprotein following cell stimulation. The hypothesis that ODC is phosphorylated at multiple sites in stimulated RAW264 cells was investigated. ODC isolated from tetradecanoyl-phorbol-13-acetate (TPA)-stimulated cells metabolically radiolabeled in the presence of $\sp{32}$P$\sb{\rm i}$ was subjected to cyanogen bromide (CNBr) cleavage followed by phosphopeptide mapping and two dimensional phosphoamino acid analysis. These phosphorylation studies demonstrated six in situ phosphorylated CNBr-generated fragments having apparent molecular weights of 17, 14.3, 8, 6.5, 4, and 2.7 kDa and also revealed that ODC is phosphorylated in RAW264 cells on at least 5 serine and 2 threonine residues.^ In addition, the in vivo specific activity and phosphorylation pattern of ODC in response to various kinase cascade stimulants was studied. A differential response in ODC specific activity and a variation in the relative distribution of $\sp{32}$P-labeling of serine and threonine residues on the ODC molecule was noted in response to fetal bovine serum, cAMP and isobutylmethylxanthine, lipopolysaccharide, or TPA.^ Based on information derived from consensus sequence motifs, three protein kinases responsible for the phosphorylation of ODC in vitro were identified. Purified ODC was phosphorylated in vitro by casein kinase II (CK II), extracellular signal-regulated kinase 1 (ERK1), and its activator, extracellular signal-regulated kinase kinase (MEK). CK II phosphorylated ODC on serine residues contained on three CNBr-generated peptides with apparent molecular weights of 14.3, 6.5, and 2.7 kDa. Both ERK1 and MEK phosphorylated ODC on serine and threonine residues on a CNBr-generated peptide fragment with an apparent molecular weight of 6.5 kDa. The in vitro radiolabeled peptides corresponded in molecular mass with some of the CNBr fragments of ODC phosphorylated in situ in stimulated RAW264 cells.^ This study concludes that ODC is phosphorylated in the transformed macrophage RAW264 cell line at multiple sites in response to various kinase cascade stimulants. These stimulants also led to a differential response in specific activity and phosphorylation pattern of ODC in RAW264 cells. Three protein kinases have been identified which phosphorylate ODC in vitro on peptides and amino acid residues which correspond with those phosphorylated in situ. ^

A FUNCTIONAL VARIANT OF HLA-DR1 DELINEATED BY T-LYMPHOCYTE CLONE REACTIVITY, PROTEIN CONFORMATION, AND GENOMIC RESTRICTION SITES

This research characterized a serologically indistinguishable form of HLA-DR1 that: (1) cannot stimulate some DR1-restricted or specific T-lymphocyte clones; (2) displays an unusual electrophoretic pattern on two dimensional gels; and (3) is marked by a polymorphic restriction site of the alpha gene. Inefficient stimulation of some DR1-restricted clones was a property of DR1$\sp{+}$ cells that shared HLA-B14 on the same haplotype and/or were carriers of 21-hydroxylase (21-OH) deficiency. Nonclassical 21-OH deficiency frequently demonstrates genetic linkage with HLA-B14;DR1 haplotypes and associates with duplications of C4B and one 21-OH gene. Cells having both stimulatory (DR1$\sb{\rm n}$) and nonstimulatory (DR1$\sb{\rm x}$) parental haplotypes did not mediate proliferation of these clones. However, heterozygous DR1$\sb{\rm x}$, 2 and DR1$\sb{\rm x}$, 7 cells were efficient stimulators of DR2 and DR7 specific clones, respectively, suggesting that a trans acting factor may modify DR1 alleles or products to yield a dominant DR1$\sb{\rm x}$ phenotype. Incompetent stimulator populations did not secrete an intercellular soluble or contact dependent suppressor factor nor did they express interleukin-2 receptors competing for T-cell growth factors. Two dimensional gel analysis of anti-DR immunoprecipitates revealed, in addition to normal DR$\alpha$ and DR$\beta$ chains, a 50kD species from DR1$\sb{\rm x}$ but not from the majority of DR1$\sb{\rm n}$ or non-DR1 cells. The 50kD structure was stable under reducing conditions in SDS and urea, had antigenic homology with DR, and dissociated after boiling into 34kD and 28kD peptide chains apparently identical with DR$\alpha$ and DR$\beta$ as shown by limited digest peptide maps. N-linked glycosylation and sialation of DRgp50 appeared to be unchanged from normal DR$\alpha$ and DR$\beta$. Bg1II digestion and $DR\alpha$ probing of DR1$\sb{\rm x}$ genomic DNA revealed a 4.5kb fragment while DR1$\sb{\rm n}$ DNA yielded 3.8 and 0.76kb fragments; all restriction sites mapped to the 3$\sp\prime$ untranslated region of $DR\alpha$. Collectively, these data suggest that DRgp50 represents a novel combinatorial association between constitutive chains of DR that may interfere with or compete for normal T cell receptor recognition of DR1 as both an alloantigen and restricting element. Furthermore, extensive chromosomal abnormalities previously mapped to the class III region of B14;DR1 haplotypes may extend into the adjacent class II region with consequent intrusion on immune function. ^

STUDIES ON THE CLASS I GLYCOPROTEINS OF THE MURINE MAJOR HISTOCOMPATIBILITY COMPLEX (TRANSPLANTATION ANTIGEN, LYMPHOCYTE, CELL SURFACE MOLECULE)

A series of studies were undertaken to analyze and compare various aspects of murine class I glycoproteins. An initial area of investigation characterized the Qa-1 alloantigens using two-dimensional gel electrophoresis. Analysis of the products of the Qa-1('b), Qa-1('c) and Qa-1('d) alleles indicated that these were distinct molecules as determined by their lack of comigration upon comparative two-dimensional gel analysis. The importance of asparagine-linked glycosylation in the cell surface expression of class I molecules was also examined. These studies employed tunicamycin, an inhibitor of N-linked glycosylation. Tunicamycin treatment of activated T lymphocytes diminished the surface expression of Qa-1 to undetectable levels; the levels of other class I molecules exhibited little or no decrease. These results indicated that N-linked glycosylation has a differential importance in the cell surface expression of various class I molecules. The molecular weight diversity of class I molecules was also investigated. Molecular weight determination of both the fully glycosylated and unglycosylated forms of H-2 and Qa/Tla region encoded molecules established that there is a significant variation in the sizes of these forms of various class I molecules. The most significant difference ((TURN)9,000 daltons) exists between the unglycosylated forms of H-2K('b) and Qa-2, suggesting that the structural organization of these two molecules may be very different. A comparative two-dimensional gel analysis of various class I glycoproteins isolated from resting and activated T and B lymphocytes indicated that class I molecules expressed on activated T cells exhibited an isoelectrophoretic pattern that was distinct from the isoelectrophoretic pattern of class I molecules expessed on the other cell populations. This difference was attributed to a lower sialic acid content of the molecules expressed on activated T cells. Analysis of cell homogenates determined that activated T cells contained a higher level of endogenous neuraminidase activity than was detected in the other populations, suggesting that this may be the basis of the lower sialic acid content. The relationship of the Qa-4 and Qa-2 alloantigens was also examined. It was established that upon mitogen activation, the expression of Qa-4 was greatly decreased, whereas Qa-2 expression was not decreased. However, an anti-Qa-2 monoclonal antibody blocked the binding of an anti-Qa-4 monoclonal antibody to resting cells. These studies established that Qa-4 is a determinant restricted to resting cells, which is closely associated on the surface with the Qa-2 molecule. ^

Molecular interactions and signal transfer between sensory rhodopsin-I and its cognate transducer

SRI is unique among known photoreceptors in that it produces opposite signals depending on the color of light stimuli. Absorption of orange light (587 nm) triggers an attractant response by the cell, whereas absorption of orange light followed by near-UV light (373 run) triggers a repellent response. Using behavioral mutants that exhibit aberrant color-sensing ability, we tested a two-conformation equilibrium model, using FRET and EPR spectroscopy. The essence of the model applied to SRI-HtrI is that the complex exists in a metastable two-conformer equilibrium which is shifted in one direction by orange light absorption (producing an attractant signal) and in the opposite direction by a second UV-violet photon (producing a repellent signal). First, by FRET we found that the E-F cytoplasmic loop of SRI moves toward the RAMP domain of the HtrI transducer during the formation of the orange-light activated signaling state of the complex. This is the first localization of a change in the physical relationship between the receptor and transducer subunits of the complex and provides a structural property of the two proposed conformers that we can monitor. Second, EPR spectra of a spin label probe at this cytoplasmic position showed shifts in the dark in the mutants toward shorter or longer EF loop-RAMP distances, explaining their behavior in terms of their mutations causing pre-stimulus shifts into one or the other conformer. ^ Next, we applied a novel electrophysiological method for monitoring the directionality of proton movement during photoactivation of SRI, to investigate the process of proton transfer in the photoactive site from the chromophore to proton acceptors on both the wildtype and aberrant color-response mutants. We observed an unexpected and critical difference in the two signaling conformations of the SRI-HtrI complex. The finding is that the vectoriality (i.e. movement away or toward the cytoplasm) of the light-induced proton transfer from the chromophore to the protein is opposite in formation of the two conformations. Retinylidene proton transfer is a common critical process in rhodopsins and these results are the first to show differences in vectoriality in a rhodopsin receptor, and to demonstrate functional importance of the direction of proton transfer. ^

Organization of the inferotemporal cortex in the macaque monkey: Connections of areas PITv and CITvp

Visual cortex of macaque monkeys consists of a large number of cortical areas that span the occipital, parietal, temporal, and frontal lobes and occupy more than half of cortical surface. Although considerable progress has been made in understanding the contributions of many occipital areas to visual perceptual processing, much less is known concerning the specific functional contributions of higher areas in the temporal and frontal lobes. Previous behavioral and electrophysiological investigations have demonstrated that the inferotemporal cortex (IT) is essential to the animal's ability to recognize and remember visual objects. While it is generally recognized that IT consists of a number of anatomically and functionally distinct visual-processing areas, there remains considerable controversy concerning the precise number, size, and location of these areas. Therefore, the precise delineation of the cortical subdivisions of inferotemporal cortex is critical for any significant progress in the understanding of the specific contributions of inferotemporal areas to visual processing. In this study, anterograde and/or retrograde neuroanatomical tracers were injected into two visual areas in the ventral posterior and central portions of IT (areas PITv and CITvp) to elucidate the corticocortical connections of these areas with well known areas of occipital cortex and with less well understood regions of inferotemporal cortex. The locations of injection sites and the delineation of the borders of many occipital areas were aided by the pattern of interhemispheric connections, revealed following callosal transection and subsequent labeling with HRP. The resultant patterns of connections were represented on two-dimensional computational (CARET) and manual cortical maps and the laminar characteristics and density of the projection fields were quantified. The laminar and density features of these corticocortical connections demonstrate thirteen anatomically distinct subdivisions or areas distributed within the superior temporal sulcus and across the inferotemporal gyrus. These results serve to refine previous descriptions of inferotemporal areas, validate recently identified areas, and provide a new description of the hierarchical relationships among occipitotemporal cortical areas in macaques. ^

Bayesian Adaptive Designs for Early Phase Clinical Trials

My dissertation focuses mainly on Bayesian adaptive designs for phase I and phase II clinical trials. It includes three specific topics: (1) proposing a novel two-dimensional dose-finding algorithm for biological agents, (2) developing Bayesian adaptive screening designs to provide more efficient and ethical clinical trials, and (3) incorporating missing late-onset responses to make an early stopping decision. Treating patients with novel biological agents is becoming a leading trend in oncology. Unlike cytotoxic agents, for which toxicity and efficacy monotonically increase with dose, biological agents may exhibit non-monotonic patterns in their dose-response relationships. Using a trial with two biological agents as an example, we propose a phase I/II trial design to identify the biologically optimal dose combination (BODC), which is defined as the dose combination of the two agents with the highest efficacy and tolerable toxicity. A change-point model is used to reflect the fact that the dose-toxicity surface of the combinational agents may plateau at higher dose levels, and a flexible logistic model is proposed to accommodate the possible non-monotonic pattern for the dose-efficacy relationship. During the trial, we continuously update the posterior estimates of toxicity and efficacy and assign patients to the most appropriate dose combination. We propose a novel dose-finding algorithm to encourage sufficient exploration of untried dose combinations in the two-dimensional space. Extensive simulation studies show that the proposed design has desirable operating characteristics in identifying the BODC under various patterns of dose-toxicity and dose-efficacy relationships. Trials of combination therapies for the treatment of cancer are playing an increasingly important role in the battle against this disease. To more efficiently handle the large number of combination therapies that must be tested, we propose a novel Bayesian phase II adaptive screening design to simultaneously select among possible treatment combinations involving multiple agents. Our design is based on formulating the selection procedure as a Bayesian hypothesis testing problem in which the superiority of each treatment combination is equated to a single hypothesis. During the trial conduct, we use the current values of the posterior probabilities of all hypotheses to adaptively allocate patients to treatment combinations. Simulation studies show that the proposed design substantially outperforms the conventional multi-arm balanced factorial trial design. The proposed design yields a significantly higher probability for selecting the best treatment while at the same time allocating substantially more patients to efficacious treatments. The proposed design is most appropriate for the trials combining multiple agents and screening out the efficacious combination to be further investigated. The proposed Bayesian adaptive phase II screening design substantially outperformed the conventional complete factorial design. Our design allocates more patients to better treatments while at the same time providing higher power to identify the best treatment at the end of the trial. Phase II trial studies usually are single-arm trials which are conducted to test the efficacy of experimental agents and decide whether agents are promising to be sent to phase III trials. Interim monitoring is employed to stop the trial early for futility to avoid assigning unacceptable number of patients to inferior treatments. We propose a Bayesian single-arm phase II design with continuous monitoring for estimating the response rate of the experimental drug. To address the issue of late-onset responses, we use a piece-wise exponential model to estimate the hazard function of time to response data and handle the missing responses using the multiple imputation approach. We evaluate the operating characteristics of the proposed method through extensive simulation studies. We show that the proposed method reduces the total length of the trial duration and yields desirable operating characteristics for different physician-specified lower bounds of response rate with different true response rates.