Automated detection of fluorescent cells in in-resin fluorescence sections for integrated light and electron microscopy.

Integrated array tomography combines fluorescence and electron imaging of ultrathin sections in one microscope, and enables accurate high-resolution correlation of fluorescent proteins to cell organelles and membranes. Large numbers of serial sections can be imaged sequentially to produce aligned volumes from both imaging modalities, thus producing enormous amounts of data that must be handled and processed using novel techniques. Here, we present a scheme for automated detection of fluorescent cells within thin resin sections, which could then be used to drive automated electron image acquisition from target regions via 'smart tracking'. The aim of this work is to aid in optimization of the data acquisition process through automation, freeing the operator to work on other tasks and speeding up the process, while reducing data rates by only acquiring images from regions of interest. This new method is shown to be robust against noise and able to deal with regions of low fluorescence.

Segmentation of phase contrast microscopy images based on multi-scale local Basic Image Features histograms.

Phase contrast microscopy (PCM) is routinely used for the inspection of adherent cell cultures in all fields of biology and biomedicine. Key decisions for experimental protocols are often taken by an operator based on typically qualitative observations. However, automated processing and analysis of PCM images remain challenging due to the low contrast between foreground objects (cells) and background as well as various imaging artefacts. We propose a trainable pixel-wise segmentation approach whereby image structures and symmetries are encoded in the form of multi-scale Basic Image Features local histograms, and classification of them is learned by random decision trees. This approach was validated for segmentation of cell versus background, and discrimination between two different cell types. Performance close to that of state-of-the-art specialised algorithms was achieved despite the general nature of the method. The low processing time ( < 4 s per 1280 × 960 pixel images) is suitable for batch processing of experimental data as well as for interactive segmentation applications.

A geometric model of defensive peripersonal space.

Potentially harmful stimuli occurring within the defensive peripersonal space (DPPS), a protective area surrounding the body, elicit stronger defensive reactions. The spatial features of the DPPS are poorly defined and limited to descriptive estimates of its extent along a single dimension. Here we postulated a family of geometric models of the DPPS, to address two important questions with respect to its spatial features: What is its fine-grained topography? How does the nervous system represent the body area to be defended? As a measure of the DPPS, we used the strength of the defensive blink reflex elicited by electrical stimulation of the hand (hand-blink reflex, HBR), which is reliably modulated by the position of the stimulated hand in egocentric coordinates. We tested the goodness of fit of the postulated models to HBR data from six experiments in which we systematically explored the HBR modulation by hand position in both head-centered and body-centered coordinates. The best-fitting model indicated that 1) the nervous system's representation of the body area defended by the HBR can be approximated by a half-ellipsoid centered on the face and 2) the DPPS extending from this area has the shape of a bubble elongated along the vertical axis. Finally, the empirical observation that the HBR is modulated by hand position in head-centered coordinates indicates that the DPPS is anchored to the face. The modeling approach described in this article can be generalized to describe the spatial modulation of any defensive response.

Important factors determining the nanoscale tracking precision of dynamic microtubule ends.

Tracking dynamic microtubule ends in fluorescence microscopy movies provides insight into the statistical properties of microtubule dynamics and is vital for further analysis that requires knowledge of the trajectories of the microtubule ends. Here we analyse the performance of a previously developed automated microtubule end tracking routine; this has been optimized for comparatively low signal-to-noise image sequences that are characteristic of microscopy movies of dynamic microtubules growing in vitro. Sequences of simulated microtubule images were generated assuming a variety of different experimental conditions. The simulated movies were then tracked and the tracking errors were characterized. We found that the growth characteristics of the microtubules within realistic ranges had a negligible effect on the tracking precision. The fluorophore labelling density, the pixel size of the images, and the exposure times were found to be important parameters limiting the tracking precision which could be explained using concepts of single molecule localization microscopy. The signal-to-noise ratio was found to be a good single predictor of the tracking precision: typical experimental signal-to-noise ratios lead to tracking precisions in the range of tens of nanometres, making the tracking program described here a useful tool for dynamic microtubule end tracking with close to molecular precision.

How reliably can paediatric professionals identify pale stool from cholestatic newborns?

BACKGROUND: The success of surgery in infants with hepatobiliary disease is inversely proportional to the age when surgery was performed. Pale stool colour is a major indicator of biliary obstruction. However, simple recognition has been inadequate, resulting in late diagnosis and referral. Objective To assess the skills of healthcare professionals in recognising pale stools. METHOD: Photographs of normal, acholic and indeterminate infant stools were shown to paediatric professionals who have regular contact with jaundiced babies at three London teaching hospitals. Each stool was classified as 'healthy' or 'suspect'. RESULTS: One-third of the stools were not correctly identified by physicians and nurses. CONCLUSION: Experienced professionals often do not recognise stool colour associated with biliary obstruction. The authors propose that stool colour cards similar to those used in Japan and Taiwan may improve early detection of hepatobiliary disease at a minimal cost.

Zen and the art of medical image registration: correspondence, homology, and quality.

Nonrigid registration (NRR) is routinely used in the study of neuroanatomy and function and is a standard component of analysis packages such as SPM. There remain many unresolved correspondence problems that arise from attempts to associate functional areas with specific neuroanatomy and to compare both function and anatomy across patient groups. Problems can result from ignorance of the underlying neurology which is then compounded by unjustified inferences drawn from the results of NRR. Usually the magnitude, distribution, and significance of errors in NRR are unknown so the errors in correspondences determined by NRR are also unknown and their effect on experimental results cannot easily be quantified. In this paper we review the principles by which the presumed correspondence and homology of structures is used to drive registration and identify the conceptual and algorithmic areas where current techniques are lacking. We suggest that for applications using NRR to be robust and achieve their potential, context-specific definitions of correspondence must be developed which properly characterise error. Prior knowledge of image content must be utilised to monitor and guide registration and gauge the degree of success. The use of NRR in voxel-based morphometry is examined from this context and found wanting. We conclude that a move away from increasingly sophisticated but context-free registration technology is required and that the veracity of studies that rely on NRR should be keenly questioned when the error distribution is unknown and the results are unsupported by other contextual information.

Biosynthesis and action of neurosteroids.

Over the past decade, it has become clear that the brain, like the gonad, adrenal and placenta, is a steroidogenic organ. However, unlike classic steroidogenic tissues, the synthesis of steroids in the nervous system requires the coordinate expression and regulation of the genes encoding the steroidogenic enzymes in several different cell types (neurons and glia) at different locations in the nervous system, and at distances from the cell bodies. The steroids synthesized by the brain and nervous system, given the name neurosteroids, have a wide variety of diverse functions. In general, they mediate their actions, not through classic steroid hormone nuclear receptors, but through other mechanisms such as through ion gated neurotransmitter receptors, or through direct or indirect modulation of other neurotransmitter receptors. We have briefly summarized the biochemistry of the enzymes involved in the biosynthesis of neurosteroids, their localization during development and in the adult, and the regulation of their expression, highlighting both similarities and differences between expression in the brain and in classic steroidogenic tissues.

Biosynthesis of the neurosteroid 3 alpha-hydroxy-4-pregnen-20-one (3 alpha hp), a specific inhibitor of FSH release.

The gonadal steroid 3 alpha-hydroxy-4-pregnen-20-one (3 alpha HP) is a neuroactive steroid with anxiolytic and analgesic actions. In addition, 3 alpha HP has been shown to inhibit GnRH activity on gonadotropes and selectively suppress FSH release from pituitary cells, without an effect on LH. The enzyme 3 alpha-hydroxysteroid dehydrogenase (3 alpha HSD) has been presumed to be the enzyme responsible for the conversion of progesterone to 3 alpha HP, but this has never been confirmed in vitro or in vivo. We have now determined the mechanism of 3 alpha HP synthesis in vivo using specific enzyme inhibitors and in vitro using recombinant proteins. Incubation of [(3)H]progesterone with purified recombinant rat and human 3 alpha HSD isoforms showed that both the rat 3 alpha HSD and the human type 2(brain) 3 alpha HSD converted progesterone to 3 alpha HP. Age-dependent 3 alpha HP production was demonstrated in pituitary and cortex. Incubation of both tissues with indomethacin, a known 3 alpha HSD inhibitor, decreased the conversion of progesterone to 3 alpha HP by at least 70%, indicating that 3 alpha HSD was responsible for this conversion. As human type 2 3 alpha HSD is expressed in a region-specific fashion in the brain, 3 alpha HP may only be made in specific regions of the brain. Furthermore, the data suggest that the pituitary has the capacity for 3 alpha HP production, which may provide an additional mechanism for regulation of GnRH action.

Selective serotonin reuptake inhibitors directly alter activity of neurosteroidogenic enzymes.

The neurosteroid 3alpha-hydroxysteroid-5alpha-pregnan-20-one (allopregnanolone) acts as a positive allosteric modulator of gamma-aminobutyric acid at gamma-aminobutyric acid type A receptors and hence is a powerful anxiolytic, anticonvulsant, and anesthetic agent. Allopregnanolone is synthesized from progesterone by reduction to 5alpha-dihydroprogesterone, mediated by 5alpha-reductase, and by reduction to allopregnanolone, mediated by 3alpha-hydroxysteroid dehydrogenase (3alpha-HSD). Previous reports suggested that some selective serotonin reuptake inhibitors (SSRIs) could alter concentrations of allopregnanolone in human cerebral spinal fluid and in rat brain sections. We determined whether SSRIs directly altered the activities of either 5alpha-reductase or 3alpha-HSD, using an in vitro system containing purified recombinant proteins. Although rats appear to express a single 3alpha-HSD isoform, the human brain contains several isoforms of this enzyme, including a new isoform we cloned from human fetal brains. Our results indicate that the SSRIs fluoxetine, sertraline, and paroxetine decrease the K(m) of the conversion of 5alpha-dihydroprogesterone to allopregnanolone by human 3alpha-HSD type III 10- to 30-fold. Only sertraline inhibited the reverse oxidative reaction. SSRIs also affected conversions of androgens to 3alpha- and 3alpha, 17beta-reduced or -oxidized androgens mediated by 3alpha-HSD type II(Brain). Another antidepressant, imipramine, was without any effect on allopregnanolone or androstanediol production. The region-specific expression of 3alpha-HSD type II(Brain) and 3alpha-HSD type III mRNAs suggest that SSRIs will affect neurosteroid production in a region-specific manner. Our results may thus help explain the rapid alleviation of the anxiety and dysphoria associated with late luteal phase dysphoria disorder and major unipolar depression by these SSRIs.

National trends in the concurrence of tuberculosis and acquired immunodeficiency syndrome.

BACKGROUND: Elucidation of the relationship between tuberculosis (TB) and the acquired immunodeficiency syndrome (AIDS) is needed to help predict the future course of these two epidemics. We examined nationwide trends in TB and AIDS occurring in the same individual. METHODS: Health departments in the 50 states, District of Columbia, Puerto Rico, and Guam matched their TB and AIDS case registries to determine the number of persons diagnosed with both TB and AIDS. The number of AIDS cases, TB cases, AIDS cases that matched with a TB case on the TB registry, and TB cases that matched with an AIDS case on the AIDS registry were reported to the Centers for Disease Control and Prevention, Atlanta, Ga. Data were analyzed for the period from 1981 through 1991. The number of matched TB-AIDS cases was compared with a modeled estimate of excess TB cases during the period from 1985 through 1990. RESULTS: From 1981 through 1991 there were 11,299 AIDS cases that matched with a TB case on the TB registry, representing 5.1% (geographic variation, 0% to 9.3%) of AIDS cases. The TB cases that matched with an AIDS case on the AIDS registry represent 4.3% (geographic variation, 0% to 15.1%) of TB cases from 1981 through 1991. Since 1981, matched TB and AIDS cases increased yearly through 1990. When examined by year of AIDS report, the percentage of AIDS cases that matched with a TB case increased from 1981 to 1982 (1.9% to 5.1%), remained fairly constant from 1983 through 1987 (range, 4.0% to 4.7%), increased in 1988 (5.4%) after extrapulmonary TB was added to the AIDS case definition, and increased slightly through 1990 (5.8%). When examined by year of TB report, the percentage of TB cases that matched with an AIDS case increased steadily from 1981 through 1990 (0.1% to 9.5%). The calculated fraction of excess TB cases during the period from 1985 through 1990 that could be accounted for by identified TB-AIDS cases was 30%. CONCLUSION: The risk of TB or AIDS among persons already diagnosed with one disease is much higher than among the general population. The percentage of persons with TB who are also diagnosed with AIDS has been increasing rapidly. Human immunodeficiency virus-induced immunosuppression is an important contributor to the TB epidemic and probably accounts for a minimum of 30% of excess TB cases during the period from 1985 through 1990.

Two mutations in the insulin receptor gene of a patient with leprechaunism: application to prenatal diagnosis.

Leprechaunism is an autosomal recessive disorder caused by mutations in the insulin receptor gene and characterized by intrauterine and postnatal growth restriction, abnormal glucose homeostasis, and severe insulin-resistance. Here we report the biochemical and molecular characterization of a male patient, NZ, who died at 2 yr of age with this syndrome. 125I-Insulin binding to fibroblasts from the proband, his mother, father, and unaffected sister was reduced to 8, 53, 38, and 35% of controls, respectively. Analysis of the insulin receptor gene by polymerase chain reaction amplification using primers flanking each of the 22 exons and direct DNA sequencing identified 2 different mutations in the proband. The paternal mutation was an in-frame deletion of base pairs 1159-1161 in exon 3, which resulted in the loss of the codon for Asn-281. The maternal mutation was a G-->A transition in the first nucleotide of the splice-donor junction in intron 13. The maternal mutation activated a cryptic splice site 27 base pairs upstream in exon 13 and caused an in-frame deletion of amino acids 859-867 of the extracellular domain of the insulin receptor beta subunit. Identification of both mutations enabled prenatal diagnosis in 2 subsequent pregnancies. In the first pregnancy, DNA from cells cultured from chorionic villus (CV) biopsies carried both mutations in the insulin receptor gene. In the second pregnancy, DNA from the CV biopsy cells was negative for both mutations, indicating that the fetus was unaffected by leprechaunism. Insulin binding could not be used in prenatal diagnosis because cells cultured from some control CV biopsies failed to bind insulin. These data indicate that patient NZ with leprechaunism was a compound heterozygote for 2 novel mutations in the insulin receptor gene and that direct DNA sequencing enables prenatal diagnosis for this lethal disorder.

Analysis of formalin-fixed and frozen myocardial autopsy samples for viral genome in childhood myocarditis and dilated cardiomyopathy with endocardial fibroelastosis using polymerase chain reaction (PCR).

Viral infection of the myocardium is implicated in the pathogenesis of myocarditis and dilated cardiomyopathy (DCM). Enteroviruses have been considered the most common viral etiologic agents, based on peripheral culture and serologic methods. Recently, polymerase chain reaction (PCR) has been shown to be useful in the detection of viral genomes from various infected organs and body fluids. In this study, myocardial samples from autopsy specimens (formalin fixed and fresh frozen) were examined for enteroviral and DNA viral (adenovirus, herpes simplex virus [HSV], and cytomegalovirus (CMV]) genome by PCR. The specimens studied were from 58 patients with myocarditis, 28 patients with DCM and endocardial fibroelastosis [EFE], and 22 controls. Viral genome was detectable in 34 of the 58 (59%) autopsy-proven myocarditis samples (18 adenovirus, 12 enterovirus, 2 CMV, 2 HSV) and 6 of the 28 samples from patients with DCM and EFE (6 adenovirus). We conclude that PCR is effective in the rapid amplification of virus from frozen and formalin-fixed myocardial samples and that adenovirus is an important etiologic agent in viral myocarditis as well as DCM with EFE.

Impaired growth in Rabson-Mendenhall syndrome: lack of effect of growth hormone and insulin-like growth factor-I.

Mutations in the insulin receptor gene cause the severe insulin-resistant syndromes leprechaunism and Rabson-Mendenhall syndrome. There is no accepted therapy for these inherited conditions. Here we report the results of recombinant human GH (rhGH) and recombinant human insulin-like growth factor-I (rhIGF-I) treatment of a male patient, Atl-2, with Rabson-Mendenhall syndrome. The patient was small for gestational age, had premature dentition, absence of sc fat, acanthosis nigricans, fasting hypoglycemia and postprandial hyperglycemia, and extremely high concentrations of circulating insulin (up to 8500 microU/mL). Fibroblasts and lymphoblasts established from this patient had reduced insulin binding, which was 20-30% of the control value. Binding of epidermal growth factor, IGF-I, and GH to the patient's fibroblasts was normal. The growth of fibroblasts cultured from patient Atl-2 in vitro was intermediate between that of fibroblasts from patients with leprechaunism and control values. The patient's growth curve in vivo was far below the fifth percentile despite adequate nutrition. To stimulate growth, therapy with rhGH was initiated, the rationale being to stimulate hepatic IGF-I production and IGF-I receptor signaling, and bypass the inherited block in insulin receptor signaling. Therapy with rhGH (up to 0.5 mg/kg.week) did not improve growth and failed to increase the levels of circulating IGF-I and IGF-binding protein-3 over a 14-month period. As rhGH could not stimulate growth, rhIGF-I (up to 100 micrograms/kg.day) was given by daily sc injection. No increase in growth velocity was observed over a 14-month period. These results indicate that both GH and IGF-I fail to correct growth in a patient with severe inherited insulin resistance. The lack of efficacy of IGF-I treatment may be related to multiple factors, such as the poor metabolic state of the patient, the deficiency of serum carrier protein for IGF-I, an increased clearance of the growth factor, IGF-I resistance in target cells at a receptor or postreceptor level, or an inhibitory action of the mutant insulin receptors on IGF-I receptor signaling.

The intrinsic geometry of the cerebral cortex.

The mammalian cerebral cortex is a profoundly convoluted six-layered surface. The expansion of the cortex during evolution appears to be due to an increase in the number of functional units as opposed to an increase in the complexity of the units. Geometric similarity predicts that cortical area should increase in proportion to the 2/3 power of cortical volume. Allometric analysis has shown that in fact cortical area increases as a nearly linear function of cortical volume. This can be understood by appreciating that smaller brains tend to be smooth (lissencephalic) and larger brains fissured (gyrencephalic). This process of fissuration has reached its modern terrestrial limit in the human brain where the majority of the cortical surface is hidden within folds. The thickness of the cortex (2-3 mm) is small compared to its area (2000-2500 cm2) so the application of the techniques of differential geometry (the mathematics of idealized surfaces) is justified. Geometric properties of surfaces fall into two categories: intrinsic properties (which are invariant under folding of the surface, e.g. distances measured on the surface) and extrinsic properties (pure folding). The extrinsic geometry of the cortex determines the anatomical appearance of the cortex and the shape of the white matter. The intrinsic curvature of the cortex affects the relative position of functional areas and the spread of activity within the surface itself. A cortical surface has been reconstructed from cross-sections. Analysis of this surface has shown that the cortex has significant intrinsic curvature and hence it is wrong to regard it as merely a crumpled bag. The particular geometry observed is such that the surface is peculiarly "close together". Theoretical considerations and simulations suggest that the intrinsic geometry may have a significant effect on: the necessity of non-uniform growth in models of cortical development; the location of integrative areas; and the synchronization of neuronal firing. It is suggested that intrinsic descriptions of the cortex may prove more natural than extrinsic ones.

Activation of glucose transport by a natural mutation in the human insulin receptor.

Naturally occurring mutations in the insulin receptor gene cause heritable severe insulin resistance. These mutations usually impair insulin receptor signaling in cells cultured from affected individuals. However, fibroblasts cultured from a patient with intrauterine growth restriction and severe insulin resistance (leprechaun Atl-1) had normal amounts of insulin receptor protein and defective insulin binding but constitutive activation of insulin-receptor autophosphorylation and kinase activity and of glucose transport. In the same fibroblasts, growth was impaired. Homozygosity for a mutation in the insulin receptor gene was suspected, since he inherited identical DNA haplotypes for this gene from both related parents. Here we report that the proband was homozygous and both parents were heterozygous for a point mutation in the insulin receptor gene converting the Arg86 codon (CGA) to Pro (CCA) (R86P). The R86P substitution is contiguous to the hydrophobic beta-sheet of the receptor alpha subunit implicated by DeMeyts et al. [DeMeyts, P., Gu, J.-L., Shymko, R. M., Kaplan, B. E., Bell, G. I. & Whittaker, J. (1990) Mol. Endocrinol. 4, 409-416] in the binding of aromatic residues of the insulin molecule. The R86P mutant insulin receptor cDNA was inserted into a plasmid under control of a simian virus 40 promoter and transfected into Chinese hamster ovary (CHO) cells. In contrast with fibroblasts from patient Atl-1, which had normal insulin receptor processing, CHO cells stably transfected with the R86P mutant cDNA (CHO-R86P) had altered posttranslational processing. The R86P mutant receptor failed to bind insulin but caused a significant increase in basal glucose transport in CHO cells. As in fibroblasts cultured from the patient, the R86P mutant insulin receptor did not stimulate growth in transfected CHO cells. These results suggest that the R86P mutation in the insulin receptor activates glucose transport without promoting cell growth and that distinct cell types process this mutant insulin receptor differently.

Activation of insulin receptor signaling by a single amino acid substitution in the transmembrane domain.

The insulin receptor is a ligand-activated tyrosine kinase composed of two alpha and two beta subunits. A single transmembrane domain composed of 23 hydrophobic residues is contained in each beta subunit. We examined the role of the transmembrane domain in regulating insulin receptor signaling by inserting a negatively charged amino acid (Asp) for Val938 (V938D). Chinese hamster ovary (CHO) cells were stably transfected with a plasmid containing both the neomycin-resistance gene and either the wild-type or the mutant (V938D) insulin receptor cDNA. Insulin binding increased similarly in CHO cells stably transfected with the wild-type and the V938D-mutant insulin receptor cDNA. Insulin stimulated glucose transport and cell growth in cells expressing the normal insulin receptor. By contrast, in the absence of insulin, glucose transport and cell growth in CHO-V938D cells were as high as in insulin-stimulated control cells and no longer responsive to insulin stimulation. Phosphorylation of the beta subunit of the insulin receptor was also increased in CHO-V938D cells not exposed to insulin. These results support an essential role of the transmembrane domain of the insulin receptor in the transduction of insulin signaling.

Characterization of human genomic yeast artificial chromosome inserts containing hexokinase 1 coding information on chromosome 10.

Hexokinase 1 (HK1) is one of four mammalian HK isoenzymes and maps to human chromosome 10. Two yeast artificial chromosomes (YACs) were identified in the Washington University human YAC library using polymerase chain reaction (PCR) primers designed with knowledge of the human HK1 cDNA sequence. YAC B129B12 is 120 kb in length and maps entirely to chromosome 10. YAC A159D5 is 400 kb in length and appears to have resulted from a recombination of chromosome 10 with non-chromosome 10 material. We report these YACs as potential resources for those interested in HK1 gene organization and mapping, as well as those desiring additional genomic information and markers on chromosome 10.

Reduced mRNA and a nonsense mutation in the insulin-receptor gene produce heritable severe insulin resistance.

Leprechaunism is an autosomal recessive syndrome of severe insulin resistance and is characterized by intrauterine growth restriction, acanthosis nigricans, hirsutism, and loss of glucose homeostasis. Here we report a new female patient of Hispanic and Afro-American descent whose fibroblasts and lymphoblasts had markedly impaired insulin binding (less than 10% of that in controls). Insulin binding to lymphoblasts established from both unrelated parents was partially impaired. Insulin-like growth factor-I (IGF-I) and epidermal growth factor (EGF) binding to the patient's fibroblasts were within the normal range. Insulin stimulation of receptor autophosphorylation and kinase activity was markedly reduced in the patient's fibroblasts. The patient's fibroblasts had both a reduced number of immunoreactive insulin receptor (6% of those in controls) and concomitantly reduced amounts of insulin-receptor mRNA, suggesting that both mutations inherited by the patient reduced insulin-receptor mRNA. Sequencing of the insulin-receptor gene and cDNA indicated that the patient was heterozygous for a paternally derived mutation at bp 1333, converting Arg372 to a STOP codon. This nonsense mutation was observed in the insulin-receptor gene, but not in cDNA, indicating reduced amounts of mRNA for the allele containing this mutation. The coding sequence of the maternally inherited insulin-receptor allele was normal. Both the marked reduction in insulin-receptor mRNA in the compound heterozygous fibroblasts of the proband and the partially reduced insulin binding in maternal cells suggest that the maternally derived mutation is located in an insulin-receptor gene sequence that controls cellular mRNA content.