A Survey of Self-Supervised and Few-Shot Object Detection

Labeling data is often expensive and time-consuming, especially for tasks such as object detection and instance segmentation, which require dense labeling of the image. While few-shot object detection is about training a model on novel (unseen) object classes with little data, it still requires prior training on many labeled examples of base (seen) classes. On the other hand, self-supervised methods aim at learning representations from unlabeled data which transfer well to downstream tasks such as object detection. Combining few-shot and self-supervised object detection is a promising research direction. In this survey, we review and characterize the most recent approaches on few-shot and self-supervised object detection. Then, we give our main takeaways and discuss future research directions. Project page: https://gabrielhuang.github.io/fsod-survey/.

Weakly supervised underwater fish segmentation using affinity LCFCN

Estimating fish body measurements like length, width, and mass has received considerable research due to its potential in boosting productivity in marine and aquaculture applications. Some methods are based on manual collection of these measurements using tools like a ruler which is time consuming and labour intensive. Others rely on fully-supervised segmentation models to automatically acquire these measurements but require collecting per-pixel labels which are also time consuming. It can take up to 2 minutes per fish to acquire accurate segmentation labels. To address this problem, we propose a segmentation model that can efficiently train on images labeled with point-level supervision, where each fish is annotated with a single click. This labeling scheme takes an average of only 1 second per fish. Our model uses a fully convolutional neural network with one branch that outputs per-pixel scores and another that outputs an affinity matrix. These two outputs are aggregated using a random walk to get the final, refined per-pixel output. The whole model is trained end-to-end using the localization-based counting fully convolutional neural network (LCFCN) loss and thus we call our method Affinity-LCFCN (A-LCFCN). We conduct experiments on the DeepFish dataset, which contains several fish habitats from north-eastern Australia. The results show that A-LCFCN outperforms a fully-supervised segmentation model when the annotation budget is fixed. They also show that A-LCFCN achieves better segmentation results than LCFCN and a standard baseline.

A Deep Learning Localization Method for Measuring Abdominal Muscle Dimensions in Ultrasound Images

Health professionals extensively use 2D US videos and images to visualize and measure internal organs for various purposes including evaluation of muscle architectural changes. US images can be used to measure abdominal muscles dimensions for the diagnosis and creation of customized treatment plans for patients with LBP, however, they are difficult to interpret. Due to high variability, skilled professionals with specialized training are required to take measurements to avoid low intra-observer reliability. This variability stems from the challenging nature of accurately finding the correct spatial location of measurement endpoints in abdominal US images. In this paper, we use a DL approach to automate the measurement of the abdominal muscle thickness in 2D US images. By treating the problem as a localization task, we develop a modified FCN architecture to generate blobs of coordinate locations of measurement endpoints, similar to what a human operator does. We demonstrate that using the TrA400 US image dataset, our network achieves a MAE of 0.3125 on the test set, which almost matches the performance of skilled ultrasound technicians. Our approach can facilitate next steps for automating the process of measurements in 2D US images, while reducing inter-observer as well as intra-observer variability for more effective clinical outcomes.

A realistic fish-habitat dataset to evaluate algorithms for underwater visual analysis

Visual analysis of complex fish habitats is an important step towards sustainable fisheries for human consumption and environmental protection. Deep Learning methods have shown great promise for scene analysis when trained on large-scale datasets. However, current datasets for fish analysis tend to focus on the classification task within constrained, plain environments which do not capture the complexity of underwater fish habitats. To address this limitation, we present DeepFish as a benchmark suite with a large-scale dataset to train and test methods for several computer vision tasks. The dataset consists of approximately 40 thousand images collected underwater from 20 habitats in the marine-environments of tropical Australia. The dataset originally contained only classification labels. Thus, we collected point-level and segmentation labels to have a more comprehensive fish analysis benchmark. These labels enable models to learn to automatically monitor fish count, identify their locations, and estimate their sizes. Our experiments provide an in-depth analysis of the dataset characteristics, and the performance evaluation of several state-of-the-art approaches based on our benchmark. Although models pre-trained on ImageNet have successfully performed on this benchmark, there is still room for improvement. Therefore, this benchmark serves as a testbed to motivate further development in this challenging domain of underwater computer vision.

Comparison of 2 CT Contrast Media Injection Systems: Visual Air Identification and Injector Face Cleaning

PURPOSE

To compare the MEDRAD Stellant and MEDRAD Stellant FLEX computed tomography contrast media injection systems in terms of radiologic technologists' ability to visually identify air in the syringes at various distances and under different lighting conditions, as well as the ease of cleaning contrast media spills on the injector faces.

METHODS

Ten experienced radiologic technologists performed 104 tests (52 per injector) in normal light and in ambient light conditions. The tests were randomized for the radiologic technologists' distance from the injector at 2 ft (61 cm), 8 ft (244 cm), and 16 ft (488 cm). In addition, the authors tested the cleaning efficiency of the Stellant injector face with raised buttons and the Stellant FLEX injector face with buttons on a flush surface by applying a mixture of contrast media and invisible ultraviolet ink. Radiologic technologists followed 2 cleaning protocols: a quick clean (5 seconds) and a full clean (1 minute). Residual contrast mixture was measured.

RESULTS

The Stellant FLEX injector had an overall higher rate of correct air identification than did the Stellant injector, 97.5% vs 86.9%, respectively (P < .001), with improvement seen at greater distances. The rates for the Stellant FLEX injector remained stable for all distances (99.4%-93.9%; P < .003). A similar result was seen with ambient lighting: The visibility of air in the Stellant FLEX injector remained stable (P < .001). During cleaning, the Stellant FLEX injector required less time to reach a greater level of cleanliness and showed less contrast mixture residue, especially after the quick clean procedure (16% and 59%, respectively; P < .001).

DISCUSSION

Injector manufacturers provide various technologies to assist radiologic technologists in visualizing the presence of air in a syringe. The Stellant injector features clear syringes with FluiDots as an air identification technology. The Stellant FLEX injector incorporates an added feature called Beacon technology. This study's results are highly clinically relevant because unintentional iatrogenic air injection occurs in clinical practice and can, in rare cases, severely harm patients.

CONCLUSION

Radiologic technologists showed an improved ability to identify air in the syringes in the Stellant FLEX system. In addition, radiologic technologists were confident about their identification, specifically at greater distances and in ambient lighting conditions. The Stellant FLEX injector face also enabled quicker and more efficient cleaning.

Intrathecal administration of autologous mesenchymal stromal cells for spinal cord injury: Safety and efficacy of the 100/3 guideline

BACKGROUND AIMS

Cell therapy with autologous mesenchymal stromal cells (MSCs) in patients with spinal cord injury (SCI) is beginning, and the search for its better clinical application is an urgent need.

METHODS

We present a phase 2 clinical trial in patients with chronic SCI who received three intrathecal administrations of 100 x 10⁶ MSCs and were followed for 10 months from the first administration. Efficacy analysis was performed on nine patients, and safety analysis was performed on 11 patients. Clinical scales, urodynamic, neurophysiological and neuroimaging studies were performed previous to treatment and at the end of the follow-up.

RESULTS

The treatment was well-tolerated, without any adverse event related to MSC administration. Patients showed variable clinical improvement in sensitivity, motor power, spasms, spasticity, neuropathic pain, sexual function or sphincter dysfunction, regardless of the level or degree of injury, age or time elapsed from the SCI. In the course of follow-up three patients, initially classified as ASIA A, B and C, changed to ASIA B, C and D, respectively. In urodynamic studies, at the end of follow-up, 66.6% of the patients showed decrease in postmicturition residue and improvement in bladder compliance. At this time, neurophysiological studies showed that 55.5% of patients improved in somatosensory or motor-evoked potentials, and that 44.4% of patients improved in voluntary muscle contraction together with infralesional active muscle reinnervation.

CONCLUSIONS

The present guideline for cell therapy is safe and shows efficacy in patients with SCI, mainly in recovery of sphincter dysfunction, neuropathic pain and sensitivity.

Cell therapy with autologous mesenchymal stromal cells in post-traumatic syringomyelia

BACKGROUND AIMS

Recently, clinical studies show that cell therapy with mesenchymal stromal cells (MSCs) improves the sequelae chronically established in paraplegic patients, being necessary to know which of them can obtain better benefit.

METHODS

We present here a phase 2 clinical trial that includes six paraplegic patients with post-traumatic syringomyelia who received 300 million MSCs inside the syrinx and who were followed up for 6 months. Clinical scales, urodynamic, neurophysiological, magnetic resonance (MR) and studies of ano-rectal manometry were performed to assess possible improvements.

RESULTS

In all the cases, MR at the end of the study showed a clear reduction of the syrinx, and, at this time, signs of improvement in the urodynamic studies were found. Moreover, four patients improved in ano-rectal manometry. Four patients improved in neurophysiological studies, with signs of improvement in evoked potentials in three patients. In the American Spinal Injury Association (ASIA) assessment, only two patients improved in sensitivity, but clinical improvement in neurogenic bowel dysfunction was observed in four patients and three patients described improvement in bladder dysfunction. Spasms reduced in two of the five patients who had them previous to cell therapy, and spasticity was improved in the other two patients. Three patients had neuropathic pain before treatment, and it was reduced or disappeared completely during the study. Only two adverse events ocurred, without relation to the cell therapy.

CONCLUSIONS

Cell therapy can be considered as a new alternative to the treatment of post-traumatic syringomyelia, achieving reduction of syrinx and clinical improvements in individual patients.

On-Board Object Detection: Multicue, Multimodal, and Multiview Random Forest of Local Experts

Despite recent significant advances, object detection continues to be an extremely challenging problem in real scenarios. In order to develop a detector that successfully operates under these conditions, it becomes critical to leverage upon multiple cues, multiple imaging modalities, and a strong multiview (MV) classifier that accounts for different object views and poses. In this paper, we provide an extensive evaluation that gives insight into how each of these aspects (multicue, multimodality, and strong MV classifier) affect accuracy both individually and when integrated together. In the multimodality component, we explore the fusion of RGB and depth maps obtained by high-definition light detection and ranging, a type of modality that is starting to receive increasing attention. As our analysis reveals, although all the aforementioned aspects significantly help in improving the accuracy, the fusion of visible spectrum and depth information allows to boost the accuracy by a much larger margin. The resulting detector not only ranks among the top best performers in the challenging KITTI benchmark, but it is built upon very simple blocks that are easy to implement and computationally efficient.

Bilingualism Influences Structural Indices of Interhemispheric Organization

Bilingualism represents an interesting model of possible experience-dependent alterations in brain structure. The current study examines whether interhemispheric adaptations in brain structure are associated with bilingualism. Corpus callosum volume and cortical thickness asymmetry across 13 regions of interest (selected to include critical language and bilingual cognitive control areas) were measured in a sample of Spanish-English bilinguals and age- and gender-matched monolingual individuals (N = 39 per group). Cortical thickness asymmetry of the anterior cingulate region differed across groups, with thicker right than left cortex for bilinguals and the reverse for monolinguals. In addition, two adjacent regions of the corpus callosum (mid-anterior and central) had greater volume in bilinguals. The findings suggest that structural indices of interhemispheric organization in a critical cognitive control region are sensitive to variations in language experience.

Structural asymmetry of the human cerebral cortex: Regional and between-subject variability of surface area, cortical thickness, and local gyrification

Structural asymmetry varies across individuals, brain regions, and metrics of cortical organization. The current study investigated regional differences in asymmetry of cortical surface area, thickness, and local gyrification, and the extent of between-subject variability in these metrics, in a sample of healthy young adults (N=200). Between-subject variability in cortical structure may provide a means to assess the extent of biological flexibility or constraint of brain regions, and we explored the potential influence of this variability on the phenotypic expression of structural asymmetry. The findings demonstrate that structural asymmetries are nearly ubiquitous across the cortex, with differing regional organization for the three cortical metrics. This implies that there are multiple, only partially overlapping, maps of structural asymmetry. The results further indicate that the degree of asymmetry of a brain region can be predicted by the extent of the region's between-subject variability. These findings provide evidence that reduced biological constraint promotes the expression of strong structural asymmetry.

Neurostructural correlates of consistent and weak handedness

Various cognitive differences have been reported between consistent and weak handers, but little is known about the neurobiological factors that may be associated with this distinction. The current study examined cortical structural lateralization and corpus callosum volume in a large, well-matched sample of young adults (N = 164) to explore potential neurostructural bases for this hand group difference. The groups did not differ in corpus callosum volume. However, at the global hemispheric level, weak handers had reduced or absent asymmetries for grey and white matter volume, cortical surface area, thickness, and local gyrification, relative to consistent handers. Group differences were also observed for some regional hemispheric asymmetries, the most prominent of which was reduced or absent gyrification asymmetry for weak handers in a large region surrounding the central sulcus and extending into parietal association cortex. The findings imply that variations in handedness strength are associated with differences in structural lateralization, not only in somatomotor regions, but also in areas associated with high level cognitive control of action.

Structural asymmetry of anterior insula: behavioral correlates and individual differences

The current study investigated behavioral correlates of structural asymmetry of the insula, and traditional perisylvian language regions, in a large sample of young adults (N=200). The findings indicated (1) reliable leftward surface area asymmetry of the anterior insula, (2) association of this asymmetry with divided visual field lateralization of visual word recognition, and (3) modulation of the correlation of structural and linguistic asymmetry by consistency of hand preference. Although leftward asymmetry of cortical surface area was observed for the anterior insula, pars opercularis and triangularis, and planum temporale, only the anterior insula asymmetry was associated with lateralized word recognition. We interpret these findings within the context of recent structural and functional findings about the human insula. We suggest that leftward structural lateralization of earlier developing insular cortex may bootstrap asymmetrical functional lateralization even if the insula is only a minor component of the adult language network.

Cortical neuronal ensembles driven by dorsal horn spinal neurones with spontaneous activity in the cat

Simultaneous recordings of cortical activity, recorded as the cortical local field potential (CLFP) in the contralateral posterior sigmoid gyrus, and the spinal activity, recorded as the cord dorsum potential (CDP) of the L6 lumbar segment, were made in the anaesthetized cat. The electrodes were positioned in somatosensory regions where the largest spontaneous negative CLFPs and CDPs were recorded. We found that spontaneous negative CLFPs were preceded by spontaneous negative CDPs with a mean latency of 14.4+/-3.5 ms. Amplitude of these spontaneous negative CLFPs was abolished after section of the dorsal columns and ipsilateral dorsolateral funiculus. It is concluded that the neurones of the primary somatosensory cortex can be driven by dorsal horn spinal neurones producing the spontaneous negative CDPs. This suggests very strongly that spontaneous neuronal activity in somatosensory regions of the brain is generated not only by ongoing activity of neurones located at supraspinal sites, but also by ongoing activity of spinal neurones.

Altered response to partial bladder outlet obstruction in mice lacking inducible nitric oxide synthase

INTRODUCTION

Following prolonged partial bladder outlet obstruction (BOO) in the mouse, cholinergic mediated detrusor contractility decreases. Previous work has demonstrated an increase in the inducible form of nitric oxide synthase (iNOS) at the mRNA and protein levels soon after obstruction. Since nitric oxide (NO), the product of the action of iNOS on molecular oxygen and l-arginine, produces vasodilation and decreases platelet aggregation, we believe it is an integral part of the initial detrusor response to obstruction. These experiments evaluated the detrusor response in mice incapable of producing iNOS.

MATERIALS AND METHODS

Wild type and knockout mice were partially obstructed for 1, 3, and 5 weeks. Physiologic evaluation consisted of cystometric analyses, and muscle strip studies in response to cholinergic and electrical stimulation. Strips were also relaxed with L-arginine, sodium nitroprusside, and 8-bromoguanosine 3' - 5' cyclic GMP, after precontraction.

RESULTS

After 5 weeks of obstruction, both wild type (WT) and knockout (KO) mouse bladders increased significantly in weight. WT bladders obstructed for 5 weeks had the greatest capacity (increase of 42%, p = 0.022), and a decreased contractile response to carbachol (decrease of 32% at 10-5 M, p = 0.018). No differences were noted at 1 and 3 weeks of obstruction. In contrast, KO mice had a significantly larger bladder capacity at 1 week of obstruction compared with WT, and had significantly lower responses to electrical stimulation than WT at the same time (p = 0.03). Additionally, after 5 weeks of obstruction, bladder capacity and contractility returned to baseline levels in KO mice, at a time when WT mice had significantly larger capacity and decreased contractility.

CONCLUSIONS

Bladder function following partial BOO in mice incapable of producing iNOS differed significantly from the normal response. Our data suggest that generation of iNOS soon after obstruction is necessary to prevent detrusor dysfunction at that time. Moreover, the enhanced function seen in KO bladders after longer periods of obstruction (5 weeks) in comparison to WT bladders suggests that reactive nitrogen species-induced protein nitrosylation may be involved in the loss of contractile function observed after more prolonged periods of obstruction.

Hypercortisolemia, hippocampal glucocorticoid receptors, and fast feedback

Glucocorticoids are critical for survival. The absence of glucocorticoids leads to the inability of the organism to cope with stress and subsequent death. However, diseases of glucocorticoid excess, such as Cushing's disease, make it clear that 'too much circulating glucocorticoids is also bad.' The need to control the circulating levels of glucocorticoids to prevent oversecretion leads to an elaborate set of checks and balances, in particular, the levels of glucocorticoid 'feedback' to the controlling elements to turn off glucocorticoid secretion and consequently maintain the homeostatic range of glucocorticoids. In order to accomplish these goals, multiple types of feedback exist which appear to use different intracellular mechanisms. In addition to multiple types of feedback, there are multiple sites of feedback including numerous limbic areas that influence the inhibition of the stress response. In this article, we will discuss the role of hippocampal limbic circuits in modulation of the stress response and the evidence which support the theory that changes in these circuits are associated with feedback abnormalities. We will review studies in humans which suggest that damage to the hippocampus can lead to changes in glucocorticoid feedback, and finally we will review the evidence that glucocorticoids themselves can lead to changes in hippocampal neurons resulting in feedback abnormalities.

Thoracic disc herniation, cord compression, and paraplegia caused by electrical injury: case report and review of the literature

The case of a patient with thoracic intervertebral disc prolapse, cord compression, and delayed paraplegia with recovery secondary to high-tension electrical current is presented. The unusual entry wound in the neck, makes this a rare event that to our knowledge has not been previously described in the literature. Magnetic resonance imaging allowed noninvasive and early diagnosis, and in the future probably will reveal the true incidence of cord compression in electrical injury victims.

Carbamazepine metabolism in humans: effect of concurrent anticonvulsant therapy

Free and total carbamazepine (CBZ) and carbamazepine-epoxide (CBZ-EP) plasma levels were obtained on 113 patients with epilepsy (18-61 years old) controlled on either monotherapy or coadministration with either phenobarbital (PB), phenytoin (PHT), valproic acid (VPA), or all three. A subset of patients were administered tetradeuterium labeled CBZ to evaluate the effects of autoinduction and coadministration of VPA on the kinetics of CBZ and its metabolite CBZ-EP. Polytherapy had variable effect on free and total CBZ plasma levels compared to monotherapy. Coadministered PHT (co-PHT), or all three anticonvulsants together (PHT, PB, and VPA: co-AEDs) decreased free and total CBZ plasma levels. No change was noted for coadministered VPA (co-VPA). Compared to monotherapy the free and total CBZ-EP levels increased with co-VPA, less with coadministered PB (co-PB), and no change with co-PHT or co-AEDs. Protein binding of CBZ and CBZ-EP was not affected by any antiepileptic drugs studied. The free and total CBZ-EP/CBZ ratio was tripled with co-VPA or co-AED's, and doubled with co-PHT or co-PB. Isotope labeling did not demonstrate any differences in half-life (t1/2), plasma clearance (Cl), or volume of distribution (Vd). Compared to naive controls, monotherapy and co-VPA decreased CBZ t1/2 by 50%, and more than doubled the CBZ Cl without a significant change in the Vd. Autoinduction is one explanation for these changes with chronic CBZ therapy.(ABSTRACT TRUNCATED AT 250 WORDS)

Anticonvulsant drugs determined by EMIT in spiked brain samples: a correlation with HPLC values

Phenobarbital (PB), diphenylhydantoin (PHT) and carbamazepine (CBZ) levels were determined in spiked rat brain homogenates by means of the EMIT assay, and these were compared with results obtained by high-pressure liquid chromatography (HPLC). The extraction procedure of the three drugs was identical for both methods, with minor modifications for EMIT (no addition of internal standard; reconstitution with the buffer used in routine plasma level analysis). Correlation coefficients between the data obtained by EMIT and HPLC were extremely good (PB, 0.90-0.98; PHT, 0.92-0.98; CBZ, 0.93-0.99). The results indicate that the EMIT system can detect PB, PHT, and CBZ in brain tissue with good accuracy, although HPLC remains the most accurate and proven method available for research purposes.

Polypeptide synthesis catalyzed by p-hydroxymercuribenzoate-modified ribosomes

The stimulation of poly(U)-directed polyphenylalanine synthesis produced by modification of Escherichia coli ribosomes with p-hydroxymercuribenzoate, at low molar ratios of reagent to ribosomes, is due to an increase in the average chain length of polyphenylalanine synthesized, and not to the activation of inactive ribosomes. At a higher molar ratio of p-hydroxymercuribenzoate to ribosomes, which produces no overall change in activity, approximately 50% of the active ribosomes present in the untreated preparation have been completely inactivated, and the remaining active ones, like the ribosomes of the stimulated preparation, synthesize polyphenylalanine at an increased rate as compared with the untreated ribosomes.

Structure of the yeast ribosomes. Proteins associated with the rRNA

Polyamines have been shown to bind to doubled stranded regions of rRNA [3]. Therefore, ribosomal proteins that can be cross linked to these molecules in the ribosomes structure must be bound to or located in the vicinity of the RNA. This technique is the first to yield results on the proteins associated with the rRNA in the eukaryotic ribosome where the lack of purified ribosomal proteins does not allow the use of direct binding studies as in bacterial systems. Proteins S7, S10, S13, S21, S22 and S27 in the small subunit and L2/3, L5, L10/12, L19/20, L22, L23, L36/37, L42 and L43' in the large subunit are labelled when cross linked to [14C]spermidine using 1,5-difluoro 2,4-dinitrobenzene and are good candidates to be RNA-binding proteins in ribosomes from Saccharomyces cerevisiae.

Inhibition, by selected antibiotics, of protein synthesis in cells growing in tissue cultures

A large number of compounds including actinobolin, adrenochrome, amicetin, anisomycin, aurintricarboxylic acid, blasticidin S, chartreusin, chlortetracycline, cycloheximide, doxycycline, edeine A1, edeine complex, emetine, fusidic acid, gougerotin, GppCH2p, oxytetracycline, pactamycin, polydextran sulphate, puromycin, pyrocatechol violet, sparsomycin and tubulosine have been tested for inhibitory effects on protein synthesis in cultured cells from both mouse fibroblasts (3T6 cells) and chick embryo fibroblasts (CEF). Essentially, similar results were obtained with both cell types with the most effective inhibitors being pactamycin, emetine, tubulosine, anisomycin and cycloheximide and with no significant inhibitory activity being detected with edeine complex, edeine A1, GppCH2p, polydextran sulphate, aurintricarboxylic acid, pyrocatechol violet and adrenochrome. The concentration of pactamycin required to produce 50% inhibition of protein synthesis approximated 5 X 10(-9) M, but for most of the inhibitors it ranged from 5 X 10(-6) M to 5 X 10(4) M. The molecular basis underlying these differences may be related, in addition to their intrinsic inhibitory power, to differences in permeability of the cells towards the various drugs tested. Alternatively, active accumulation of the drugs by the cells may be the variable parameter.

Proteins associated with rRNA in the Escherichia coli ribosome

Ribosomal proteins located near the rRNA have been identified by cross linking to [14C]spermine with 1,5-difluoro-2,4-dinitrobenzene. The polyamine binds to double-stranded rRNA; those proteins showing radioactivity covalently bound after treatment with the bifunctional reagent should therefore be located in the vicinity of these regions of rRNA. Six proteins from the small subunit, S4, S5, S9, S18, S19 and S20 and ten proteins from the large subunit L2, L6, L13, L14, L16, L17, L18, L19, L22 and L27 preferentially take up the label. The results obtained with three proteins from the large subunit, L6, L16 and L27, show a high degree of variability that could reflect differences of conformation in the subunit population. Several proteins were drastically modified by the cross-linking agent but were not detected in the two-dimensional gel electrophoresis (e.g., S1, S11, S21, L7, L8 and L12) and therefore could not be studied.

Binding of [3H]narciclasine to eukaryotic ribosomes. A study on a structure-activity relationship

[3H]Narciclasine is a specific inhibitor of peptide bond formation on eukaryotic ribosomes and binds to 60-S ribosomal subunits. Binding of [3H]-narciclasine to yeast ribosomes is inhibited by many other inhibitors of peptide bond formation including anisomycin, several sequiterpene antibiotics (trichodermin, trichothecin, fusarenon X and verrucarin A) several Cephalotaxus alkaloids (harringtonine, homoharringtonine and isoharringtonine), several Amaryllidaceae alkaloids (pretazettine, haemanthamine, lycorine, pseudolycorine and dihydrolycorine) and the narciclasine derivatives trans-dihydronarciclasine, trans-dihydronarciclasine acetonide and isonarciclasine. Binding is also inhibited, although to a very small extent, by methylnarciclasine and cisdihydronarciclasine. In contrast, no inhibition of [3H]narciclasine binding was observed in the presence of certain other inhibitors of peptide bond formation including blasticidin S, gougerotin, sparsomycin and puromycin.

Competition between trichodermin and several other sesquiterpene antibiotics for binding to their receptor site(s) on eukaryotic ribosomes

1. Of the five sesquiterpene antibiotics tested and found to inhibit protein synthesis in yeast spheroplasts, trichothecin, trichodermol or trichodermin stabilized polyribosomes whereas, in contrast, verrucarin A or T-2 toxin induced 'run off' of polyribosomes with a corresponding increase in 80S monoribosomes. The effect of fusarenon X on the system could not be determined as the drug failed to enter the cells. 2. [acetyl-14C]Trichodermin bound to yeast polyribosomes with a dissociation constant of 2.10 muM and to yeast 'run off' ribosomes with a dissociation constant of 0.72 muM. 3. Trichothecin, trichodermol, fusarenon X, T-2 toxin and verrucarin A competed with [acetyl-14C]trichodermin for binding to its receptor site on 'run off' ribosomes. The observed competition was quantitatively similar for all drugs tested. In contrast, the five drugs competed to different extents with trichodermin for binding to its receptor site on polyribosomes. Thus trichothecin competed with relative efficiency, whereas verrucarin A competed poorly, and the other drugs occupied intermediate positions between these two extremes. 4. Studies were also carried out with yeast 'run off' ribosomes prepared from both a wild-type strain and a strain resistant to trichodermin. Competition experiments between verrucarin A and [3H]anisomycin indicated that verrucarin A bound to 'run off' ribosomes from the mutant strain less efficiently than to those from the wild-type.

Activities of protein-deficient particles derived from 50-S ribosomal subunits by NH4Cl/ethanol treatment

Protein-deficient ribosomal particles obtained by treatment of 50-S subunits from Escherichia coli ribosomes with 1 M NH4Cl and 50% ethanol contain less than 3% of proteins L7 and L12 and about 7% of proteins L10 and L11. Proteins L1, L5, L8/9 and L25 are also released during the treatment but in amounts accounting for less than 40%. The particles are able to form peptide bonds in different systems, such as 'fragment reaction', puromycin reaction and formation of dipeptides. They also bind N-acetylphenylalanyl-tRNA and phenylalanyl-tRNA non-enzymically but are unable to support any of the elongation-factor-dependent reactions tested. However, when methanol is present, they display up to 20% of the control EF-G-dependent GTP activities such as GTP hydrolysis and formation of the ternary complex EF-G-GuoPP(CH2)P-ribosome. The first activity is totally sensitive to the antibiotic thiostrepton while the formation of the ternary complex is unaffected by the drug. When measured by equilibrium dialysis the core particles are shown to be able to bind radioactive thiostrepton. The results show that protein L11 is not an absolute requirement either for peptidyl transferase activity or for the binding of thiostrepton, although in the last case the protein strongly enhances the ribosome affinity for the antibiotic.

Initiation of the polypeptide chain by reticulocyte cell-free systems. Survey of different inhibitors of translation

In order to elucidate the mechanism of action of inhibitors that block the initiation of protein synthesis in mammalian systems, we have studied the following steps: (a) formation of the ternary complex Met-tRNAr-IF-E2-GTP, (b) binding of the initiator Met-tRNAf to the 40-S ribosomal subunit in the presence of initiation factors and dependent or not on the addition of mRNA, (c) formation of the initiation complex with 80-S ribosomes and (d) formation of the first peptide bond. Adrenochrome, aurintricarboxylic acid, polydextran sulphate, pyrochatechol violet and showdomycin block the formation of the ternary complex Met-tRNAf-IF-E2-GTP. Edeine A1, aurintricarboxylic acid and polydextran sulphate block the binding of the mRNA to the 40-S ribosomal subunit. Pactamycin induces the formation of stable smaller initiation complexes which are unable to go through the subsequent steps of initiation. Stimulation of the binding of the initiator Met-tRNAf to the 80-S ribosome in the presence of initiation factors is observed with sparsomycin and antibiotics of the sesquiterpene family (verrucarin A, trichodermin and trichothecin). However, these antibiotics block the reaction of the bound Met-tRNAf with puromycin. Narciclasine has no effect on the binding of the initiator to the ribosome but strongly blocks its reaction with puromycin. We have developed a simple technique to detect the Met-tRNAf-40-S-subunit-poly(A, G, U) initiation complexes by chromatography on Sepharose 6B columns. The requirements for the formation of such complexes measured by this technique and its comparison with the sucrose gradient centrifugation method are described.

Structure and function of rat-liver ribosomes. Modification by 2-methoxy-5-nitrotropone treatment

Rat liver ribosomes and 60-S ribosome subunits were treated with the primary-amino-group-specific reagent 2-methoxy-5-nitrotropone. Important differences in the sensitivity of several ribosomal activities to inactivation by the reagent were observed. While elongation-factor-dependent activities are totally abolished in the treated particles, peptidyl transferase activity is either unaffected in 60-S ribosomal subunits or even strongly stimulated in 80-S ribosomes. Analysis of the ribosomal proteins modified by nitrotropone made it possible to draw some conclusions on their accessibility in the ribosomal structure and to relate some proteins with their involvement in the ribosome active centers. Thus, proteins L3, L13, L15 and L23 seem to be in a rather well protected position while proteins L10, L35, L37, X1 and X2 are totally exposed to the reagent. The protein accessibility also depends on the ribosome conformation, proteins L14 and L17, for instance, being sensitive in 80-S ribosomes and protected in 60-S subunits. In relation to the implication of proteins in functional centers, the data presented here together with other data obtained from protein-deficient core particles seem to indicate a possible role of proteins L21 and/or L26 in the peptidyl transferase center.

Polypeptide-chain elongation promoted by guanyl-5'-yl imidodiphosphate

In a purified system from Escherichia coli containing ribosomes complexed with poly(uridylic acid) and N-acetyl-phenylalanyl-tRNA, the nonhydrolyzable analog of GTP, guanyl-5'-yl imidodiphosphate (Guo-5'-P2-NH-P), promotes polypeptide synthesis at a rate several times slower than GTP. The activity is completely dependent on elongation factors EF-T (i.e, EF-Ts + EF-Tu) and EF-G. Examination of individual steps of the elongation cycle in partial reactions shows that Guo-5'-P2-NH-P is as efficient as GTP in promoting the EF-T-dependent binding of phenylalanyl-tRNA to the ribosomal A site. In contrast, Guo-5'-P2-NH-P promotes the translocation-dependent binding of phenylalanyl-tRNA to a ribosome complexed with A-site-bound N-acetyl-phenylalanyl-tRNA much more slowly than GTP. This slow rate of binding is due to the presence of EF-G on the ribosome, and not to sluggish translocation, since (a) the rate remains slow even after translocation of N-acetylphenylalanyl-tRNA is completed, (b) it is greatly speeded up by removal of EF-G from the reaction mixture (after translocation has occurred), and (c) it is slowed down again by readdition of the factor. Moreover, with post-translocated ribosomes and in the absence of EF-G, formation of dipeptide subsequent to the EF-T-dependent binding of phenylalanyl-tRNA is much slower when binding of this substrate has been promoted by Guo-5'-P2-NH-P than it is when promoted by GTP. The results suggest that, during polymerization with Guo-5'-P2-NH-P, EF-G and EF-Tu are slowly released from the ribosome and, consequently, the steps of the elongation cycle subsequent to translocation and aminoacyl-tRNA binding (aminoacyl-tRNA binding and peptide bond formation, respectively) are delayed. Thus, durong elongation cycle, GTP hydrolysis is probably essential for fast release of the factors from the ribosome.

Activities of nucleoprotein particles derived from rat liver ribosome

80-S ribosomes and 60-S subunits from rat liver were treated at increasing KC1 concentrations giving protein-deficient ribosomal particles whose components were analyzed and their activity tested. Most of the activities assayed stand treatment up to KC1 concentrations of around 0.6 M; peptidyl transferase, measured by the fragment reaction, however was 50% inhibited by 0.5 M KC1 in 60-S subunits but not in 80-S ribosomes. Three proteins, L21, L26 and L31, might be implicated in this loss of activity. 60-S subunits forming part of the 80 S ribosome are more resistant to the salt treatment and the pattern of proteins released by the treatment differs from the one obtained from free 60-S subunits, implying perhaps a change of conformation of this subunit upon association to form 80-S couples. According to their resistance to release by KC1 the proteins of the large sub-unit can be divided into three groups: (1) easily removed, including proteins: L1, L11, L17 and L25 in 80-s subunits and in addition, L5, L8, L9, L13, L20, L22, L26, L29, L31 and L32/33 in 60-S subunits; (2) proteins resistant to release by high salt concentrations in 80-S ribosomes as well as in 60-S subunits, namely proteins L3, L14, L27, L36, L40, L41, X1 and X2; (3) the rest of the proteins which are released in a more or less continuous way throughout the treatment. 5 S RNA is not released by KC1 treatment at the concentrations used. The binding sites for the antibiotics trichodermin and anisomycin are affected in a different way by the salt treatment, indicating that they are structurally different.

Binding of aminoacyl-tRNA to rat liver ribosomal proteins

Rat liver ribosome treatment with ethanol and 1 M NH4Cl releases some 31-33 ribosomal proteins. This split protein fraction binds Phe-tRNA, Ac-Phe-tRNA, Met-tRNAM and f-Met-tRNAF in the absence of K+ and Mg++ ions. When the split protein fraction is passed through Sephadex G-100 only six proteins are retained in the column: S10, S14, S15, S19, L35, and L36. The aminoacyl-tRNA binding activity of this protein fraction retained in the Sephadex G-100 column is similar to that of the total split protein fraction, suggesting that the above six proteins, or only some of them, are involved in the binding reaction.

A form of elongation factor G insensitive to N-ethyl-maleimide

Treatment of elongation factor G (EF-G) with the thiol reagent N-ethylmaleimide only partially inhibits (10 to 70%) the activity of the factor in (a) guanosine nucleotide-EF-G-ribosome complex formation, (b) uncoupled ribosome-dependent GTP hydrolysis, and (c) polypeptide synthesis. Moreover, a similar treatment of the factor with N-[3H]ethylmaleimide does not lead to 3H-label being associated with a GDP-EF-G-ribosome-fusidic acid complex. Thus, the results indicate the presence in EF-G preparations of a form of the factor that does not react with N-ethylmaleimide.

Stoichiometry of polypeptide chain elongation

To quantitate the amount of GTP hydrolyzed during polypeptide chain elongation, an in vitro system containing purified endogenous Escherichia coli polysomes has been developed. The polysomes are washed with 1 M NH4Cl to eliminate endogenous GTPase activities and are depleted of subunits and free ribosomes to diminish the uncoupled elongation factor G-dependent GTP hydrolysis. These polysomes, supplemented with elongation factors, aminoacyl-tRNA, and low concentrations of GTP, incorporate amino acids in their nascent peptide chains. After correcting for a background of uncoupled GTP hydrolysis, it has been found that the incorporation of each molecule of amino acid is associated with the hydrolysis of 2 molecules of GTP.

Inhibitors of protein synthesis in eukarytic cells. Comparative effects of some amaryllidaceae alkaloids

The effects of eighteen compounds obtained from bulbs of the Amaryllidaceae family were tested on (a) animal cell growth, (b) DNA, RNA and protein synthesis by intact cells and (c) protein synthesis in cell-free systems. Dihydrolycorine, haemanthamine, lycorine, narciclasine, pretazettine and pseudolycorine halted HeLa cell growth at 10(-1) mM or lower concentrations. These compounds at their growth inhibitory concentrations block protein synthesis in ascites cells and stabilize HeLa cell polysomes in vivo. Endomyocarditis virus RNA-directed cell-free polypeptide synthesis by an ascites S-30 extract and acetyl-[14C]leucyl-puromycin formation by ascites ribosomes are also inhibited by the six compounds indicated above. It is therefore concluded that they halt protein synthesis in eukaryotic cells by inhibiting the peptide bone formation step.