Automatic worm detection to solve overlapping problems using a convolutional neural network.

The nematode Caenorhabditis elegans is a powerful experimental model to investigate vital functions of higher organisms. We recently established a novel method, named "pond assay for the sensory systems (PASS)", that dramatically improves both the evaluation accuracy of sensory response of worms and the efficiency of experiments. This method uses many worms in numbers that are impractical to count manually. Although several automated detection systems have been introduced, detection of overlapped worms remains difficult. To overcome this problem, we developed an automated worm detection system based on a deep neural network (DNN). Our DNN was based on a "YOLOv4″ one-stage detector with one-class classification (OCC) and multi-class classification (MCC). The OCC defined a single class for worms, while the MCC defined four classes for the number of overlapped worms. For the training data, a total of 2000 model sub-images were prepared by manually drawing square worm bounding boxes from 150 images. To make simulated images, a total of 10-80 model images for each class were randomly selected and randomly placed on a simulated microscope field. A total of 19,000 training datasets and 1000 validation datasets with a ground-truth bounding-box were prepared. We evaluated detection accuracy using 150 images, which were different from the training data. Evaluation metrics were detection error, precision, recall, and average precision (AP). Precision values were 0.91 for both OCC and MCC. However, the recall value for MCC (= 0.93) was higher than that for OCC (= 0.79). The number of detection errors for OCC increased with increasing the ground truth; however, that for MCC was independent of the ground truth. AP values were 0.78 and 0.90 for the OCC and the MCC, respectively. Our worm detection system with MCC provided better detection accuracy for large numbers of worms with overlapping positions than that with the OCC.

Pond Assay for the Sensory Systems of Caenorhabditis elegans: A Novel Anesthesia-Free Method Enabling Detection of Responses to Extremely Low Chemical Concentrations.

Chemotaxis in the nematode Caenorhabditis elegans has basically been examined using conventional assay methods. Although these can be problematic, for example, in their use of anesthesia, the method has never been improved. We propose a pond assay for the sensory systems (PASS) of C. elegans as a novel population-based method of behavioral analysis. The test solution is injected into a recess(es) formed on agar and the response of C. elegans to its odor and/or taste is examined. Once C. elegans individuals fall into recesses (ponds) filled with liquid, they cannot return to a solid medium. In this way, the animals are trapped with certainty without the use of anesthesia. The anesthesia used to keep animals in the attractant area in conventional chemotaxis assays is no longer required, allowing pure evaluation of the attractant or repellent response to specific substances. Furthermore, the assay itself can be greatly streamlined because the preparation can be completed simply by providing a recess(es) and filling the liquid. The present paper reports the detailed method and effectiveness of the novel PASS.

Knockdown of COPA, identified by loss-of-function screen, induces apoptosis and suppresses tumor growth in mesothelioma mouse model.

Malignant mesothelioma is a highly aggressive tumor arising from serosal surfaces of the pleura and is triggered by past exposure to asbestos. Currently, there is no widely accepted treatment for mesothelioma. Development of effective drug treatments for human cancers requires identification of therapeutic molecular targets. We therefore conducted a large-scale functional screening of mesothelioma cells using a genome-wide small interfering RNA library. We determined that knockdown of 39 genes suppressed mesothelioma cell proliferation. At least seven of the 39 genes-COPA, COPB2, EIF3D, POLR2A, PSMA6, RBM8A, and RPL18A-would be involved in anti-apoptotic function. In particular, the COPA protein was highly expressed in some mesothelioma cell lines but not in a pleural mesothelial cell line. COPA knockdown induced apoptosis and suppressed tumor growth in a mesothelioma mouse model. Therefore, COPA may have the potential of a therapeutic target and a new diagnostic marker of mesothelioma.

A loss of function screen identifies nine new radiation susceptibility genes.

Genomic instability is considered a hallmark of carcinogenesis, and dysfunction of DNA repair and cell cycle regulation in response to DNA damage caused by ionizing radiation are thought to be important factors in the early stages of genomic instability. We performed cell-based functional screening using an RNA interference library targeting 200 genes in human cells. We identified three known and nine new radiation susceptibility genes, eight of which are linked directly or potentially with cell cycle progression. Cell cycle analysis on four of the genes not previously linked to cell cycle progression demonstrated that one, ZDHHC8, was associated with the G2/M checkpoint in response to DNA damage. Further study of the 12 radiation susceptibility genes identified in this screen may help to elucidate the molecular mechanisms of cell cycle progression, DNA repair, cell death, cell growth and genomic instability, and to develop new radiation sensitizing agents for radiotherapy.

Identification of the XPG region that causes the onset of Cockayne syndrome by using Xpg mutant mice generated by the cDNA-mediated knock-in method.

In addition to xeroderma pigmentosum (XP), mutations in the human XPG gene cause early onset of Cockayne syndrome (CS) in some patients (XPG/CS). The CS-causing mutations in such patients all produce truncated XPG proteins. To test the hypothesis that the CS phenotype, with characteristics such as growth retardation and a short life span in XPG/CS patients, results from C-terminal truncations, we constructed mutants with C-terminal truncations in mouse XPG (Xpg) (from residue D811 to the stop codon [XpgD811stop] and deletion of exon 15 [Xpg Delta ex15]). In the XpgD811stop and Xpg Delta ex15 mutations, the last 360 and 183 amino acids of the protein were deleted, respectively. To generate Xpg mutant mice, we devised the shortcut knock-in method by replacing genomic DNA with a mutated cDNA fragment (cDNA-mediated knock in). The control mice, in which one-half of Xpg genomic DNA fragment was replaced with a normal Xpg cDNA fragment, had a normal growth rate, a normal life span, normal sensitivity to UV light, and normal DNA repair ability, indicating that the Xpg gene partially replaced with the normal cDNA fragment retained normal functions. The XpgD811stop homozygous mice exhibited growth retardation and a short life span, but the Xpg Delta ex15 homozygous mice did not, indicating that deletion of the last 360 amino acids results in the CS phenotype but deletion of the last 183 amino acids does not. The XpgD811stop homozygous mice, however, exhibited a slightly milder CS phenotype than did the Xpg null mutant mice, indicating that the XpgD811stop protein still retains some Xpg function that affects the severity of the CS phenotype.

p75 Is Required for the Establishment of Postnatal Sensory Neuron Diversity by Potentiating Ret Signaling.

Producing the neuronal diversity required to adequately discriminate all elements of somatosensation is a complex task during organogenesis. The mechanisms guiding this process during dorsal root ganglion (DRG) sensory neuron specification remain poorly understood. Here, we show that the p75 neurotrophin receptor interacts with Ret and its GFRα co-receptor upon stimulation with glial cell line-derived neurotrophic factor (GDNF). Furthermore, we demonstrate that p75 is required for GDNF-mediated Ret activation, survival, and cell surface localization of Ret in DRG neurons. In mice in which p75 is deleted specifically within sensory neurons beginning at E12.5, we observe that approximately 20% of neurons are lost between P14 and adulthood, and these losses selectively occur within a subpopulation of Ret+ nonpeptidergic nociceptors, with neurons expressing low levels of Ret impacted most heavily. These results suggest that p75 is required for the development of the nonpeptidergic nociceptor lineage by fine-tuning Ret-mediated trophic support.

Severe growth retardation and short life span of double-mutant mice lacking Xpa and exon 15 of Xpg.

In addition to xeroderma pigmentosum (XP), mutations in the human XPG gene cause an early onset of Cockayne syndrome (CS) in some patients (XP-G/CS) with characteristics, such as growth retardation and a short life span. In the previous studies, we generated four Xpg mutant mice with two different C-terminal truncations, null, or a base substitution mutation to identify the protein region that causes the onset of CS, and found that the CS-causing mutations, null or a deletion of the last 360 amino acids, completely inhibited the NER activity of mouse XPG (Xpg), but the non-CS-causing mutations, XpgD811A (base substitution that eliminates the nuclease activity of Xpg) or XpgDeltaex15 (deletion of the exon 15 corresponding to the last 183 amino acids), resulted in the retention of residual NER activity. To understand why mutations that completely eliminate the NER activity of Xpg cause CS but those that abolish the nuclease activity without totally eliminating the NER activity of Xpg do not result in CS, we made a series of Xpg mutant mice with Xpa-null mutant allele and found that mice with the non-CS-causing deletion mutation (XpgDeltaex15) exhibited the CS phenotype when XPA was also absent but the base substitution mutation (XpgD811A) that eliminated the Xpg nuclease activity did not. These results indicate that Xpg has a second function, beside NER, and that the disruption of this second function (deletion of the last 183 amino acids) when combined with an NER defect causes CS. When we compared amino acid sequences corresponding to the exon 15 of Xpg, a significant homology was conserved among vertebrates, but not in Drosophila and Saccharomyces cerevisiae. These observations suggest that the second function of XPG may be conserved only in vertebrates and CS symptoms may occur in its absence.

Fractionated irradiation augments inter-strain variation of skin reactions among three strains of mice.

The multifraction regimens commonly used in conventional clinical radiotherapy are largely based on radiobiological experiments. However, no experimental reports on skin reactions focusing on inter-strain differences have displayed clinical relevance to the fractionated dose schedule. In this study, mice of inbred strains A/J, C57BL/6J, and C3H/HeMs were used to reveal inter-strain difference after multifractionated irradiation. Irradiation was performed daily at graded doses of 30-60 Gy total doses, with 10 fractions of 3-6 Gy. Acute skin reactions following irradiation were scored for 50 days after irradiation. Dividing a dose into a number of fractions obviously spared skin damage in the three strains of mice. No mouse exhibited a skin damage score more than 1.5, while single dose irradiation resulted in skin damage scores up to 3. The three different strains, however, showed varying susceptibility to fractionated irradiation within the range under 1.5. C3H/HeMs did not display any skin reaction after irradiation with 40 Gy total dose, while C57BL/6J and A/J demonstrated various skin reactions. Different latent periods of damage were also observed among the strains after irradiation at each dose. Our data suggest that genetic factors cause obvious variations in severity of damage and latent period after fractionated irradiation.

Radiosensitivity of peripheral blood lymphocytes obtained from patients with cancers of the breast, head and neck or cervix as determined with a micronucleus assay.

The in vitro radiation sensitivities of peripheral blood lymphocytes obtained from 48 normal females and 168 female cancer patients were measured with the cytokinesis-blocking micronucleus assay. Cancer patients group had significantly higher mean baseline micronucleus frequency than normal healthy controls. Breast cancer patients were more radiosensitive than normal individuals. Cervical cancer cases were less radiation sensitive than normal subjects. The relative lack of radiation sensitivity in cervical-cancer cases could be due to modification of the radiosensitivity of patients' immune-responsible cells by human papillomaviruses infection. Normal individuals and cancer patients were classified according to their radiation sensitivity which was evaluated with the radiation-induced micronucleus frequencies. Such a classification will be an important initial step to characterize the radiosensitive, radioresistant, or cancer-prone individuals using specific SNP typing.

Different radiation susceptibility among five strains of mice detected by a skin reaction.

Published reports about skin reactions to radiotherapy, especially among breast-cancer patients, suggest that there are interindividual differences in the normal tissue response, and genetic factors are thought to be involved in this variation. An analysis of murine strain differences may reveal the mechanism of genetic factors in the extent of normal tissue damage from irradiation for several endpoints. The variation in the radiation susceptibility was observed when the skin of mice from strains A/J, C3H/HeMs, C57BL/6J, C.B.17/Icr-scid and C3H-scid was irradiated with a single dose ranging from 10 to 60 Gy, using Cs-137 gamma rays. The active skin reaction of A/J mice lasted for months. C3H/HeMs mice showed dose-dependent skin damage, and consequently recovered to a state of mild damage within 40 days after local irradiation. The time course of the response in C57BL/6J mice was shorter than in A/J mice. The 2 strains of scid mice exhibited severe damage after irradiation at any dose from 20 to 50 Gy, and did not show any dose dependency. The variation between murine strains in macroscopic and histopathological changes in skin during the progression and resolution of damage caused by irradiation suggests an inter-strain variation in the expression of genes involved in injury, apoptosis, repair, and remodeling.

A genetic mouse model carrying the nonfunctional xeroderma pigmentosum group G gene.

A genetic mouse model with a disrupted XPG allele was generated by insertion of neo cassette sequences into exon 3 of the XPG gene by using embryonic stem (ES) cell techniques. The xpg-deficient mice showed distinct developmental characteristics. Their body was marked smaller than that in wild-type littermates since the postnatal day 6, and this postnatal growth failure became more severe with developmental proceeding. Their life span was very short, all of the mutants died by postnatal day 23 after showing great weakness and emaciation. In addition, the mutant homozygous mice also showed some progressive neurological signs, like the lower level of activity and a progressive ataxia. Further examination indicated there was developmental retardation of the brain in the mutant mice. Their brain weight, and thickness of cerebral cortex and cerebellar cortex were significant different from the controls. These characteristics, like small size brain, brain developmental retardation and progressive neurological dysfunctions in the homozygotes were similar to the typical clinical phenotype of the XPG patients and Cockayne syndrome, we believe that the xpgdeficient mice will be an animal model for studying the function of the XP-G protein in nucleotide-excision repair and mechanisms related to the clinic symptoms of XP-G and Cockayne syndrome in humans.

Comparison of conventional and novel PET tracers for imaging mesothelioma in nude mice with subcutaneous and intrapleural xenografts.

INTRODUCTION: Malignant mesothelioma is a highly aggressive tumor originating in the pleura, peritoneum and pericardium, and the prognosis of patients undergoing current treatment remains poor. To develop new therapies, it is important to have a noninvasive imaging system for evaluating the efficacy of such prospective treatments. We have established clinically relevant mouse models and evaluated conventional and novel positron emission tomography (PET) tracers. METHODS: Epithelioid and sarcomatoid mesothelioma cells were inoculated subcutaneously and intrapleurally into nude mice. Biodistribution and PET imaging studies were conducted by injecting [(18)F]fluoro-2-deoxy-D-glucose (FDG), 3'-[(18)F]fluoro-3'-doxythymidine (FLT) or 4'-methyl-[(11)C]thiothymidine (S-dThd) into the mouse models. In vitro cellular uptake of [(14)C]FDG and [(3)H]FLT and thymidine kinase 1 (TK(1)) activity in both cell lines were measured. Expression of glucose transporter 1 (GLUT-1) and Ki-67 in xenografted tumors was evaluated by immunohistochemical staining. RESULTS: In epithelioid mesothelioma models, biodistribution experiments showed that tumor uptake of [(11)C]S-dThd was significantly higher than that of [(18)F]FDG. On the other hand, in sarcomatoid models, [(18)F]FDG showed significantly higher accumulation than the other two tracers. These differential uptakes of the three tracers were confirmed by PET imaging. The cellular uptake of [(14)C]FDG and [(3)H]FLT and TK(1) activity in sarcomatoid cells were higher than those of epithelioid cells. GLUT-1 protein was strongly expressed in sarcomatoid but not in epithelioid tumor. We observed a high percentage of Ki-67-positive cells in both epithelioid and sarcomatoid tumors. CONCLUSIONS: We established nude mouse models of epithelioid and sarcomatoid subtypes of mesothelioma. PET tracers applicable for the evaluation of epithelioid and sarcomatoid mesothelioma would vary: [(18)F]FLT and [(11)C]S-dThd seemed suitable for the epithelioid subtype and [(18)F]FDG seemed suitable for the sarcomatoid subtype in our mouse models. Our results indicated that cellular uptake and TK(1) activity in vitro are not always consistent with tracer uptake of [(18)F]FLT and [(11)C]S-dThd in vivo. These mouse models and PET imaging might be useful tools for evaluating new and effective treatments in mesothelioma.

Defective repair of radiation-induced DNA damage is complemented by a CHORI-230-65K18 BAC clone on rat chromosome 4.

The Long Evans cinnamon (LEC) rat is highly susceptible to X-irradiation due to defective DNA repair and is thus a model for hepatocellular carcinogenesis. We constructed a bacterial artificial chromosome (BAC) contig of rat chromosome 4 completely covering the region associated with radiation susceptibility. We used transient and stable transfections to demonstrate that defective DNA repair in LEC cells is fully complemented by a 200-kb BAC, CHORI-230-65K18. Further analysis showed that the region associated with radiation susceptibility is located in a 128,543-bp region of 65K18 that includes the known gene Rpn1. However, neither knockdown nor overexpression of Rpn1 indicated that this gene is associated with radiation susceptibility. We also mapped three ESTs (TC523872, TC533727, and CB607546) in the 128,543-bp region, suggesting that 65K18 contains an unknown gene associated with X-ray susceptibility in the LEC rat.

Inter-strain variance in late phase of erythematous reaction or leg contracture after local irradiation among three strains of mice.

AIM: To gain insights into inter-strain differences in radiosensitivity. METHODS: Mice of inbred strains, A/J, C57BL/6J, and C3H/HeMs, were irradiated at graded doses ranging from 20 to 60 Gy. Skin reaction and leg contraction were observed for a period of 230 days and between 175 and 350 days, respectively. Gene expressions in leg skin tissue were quantified by quantitative RT-PCR assay at 1, 12 and 72 h after 30 Gy irradiation. Mice were locally irradiated by using a Cs-137 source. RESULTS: The three strains showed various degrees of susceptibility to irradiation has evaluated by skin scores. Large inter-strain differences were also detected in the lengths of contraction. Expressions of several genes such as Per3 and Rad51ap1 displayed inter-strain differences. CONCLUSIONS: The continuum model of tissue injury revealed that genetic factor, which varies among strains, is one of the causes of variances in severity of damage after irradiation.

A fast, simple method for screening radiation susceptibility genes by RNA interference.

Radiotherapy can cause unacceptable levels of damage to normal tissues in some cancer patients. To understand the molecular mechanisms underlying radiation-induced physiological responses, and to be able to predict the radiation susceptibility of normal tissues in individual patients, it is important to identify a comprehensive set of genes responsible for radiation susceptibility. We have developed a simple and rapid 96-well screening protocol using cell proliferation assays and RNA interference to identify genes associated with radiation susceptibility. We evaluated the performance of alamarBlue-, BrdU-, and sulforhodamine B-based cell proliferation assays using the 96-well format. Each proliferation assay detected the known radiation susceptibility gene, PRKDC. In a trial screen using 28 shRNA vectors, another known gene, CDKN1A, and one new radiation susceptibility gene, ATP5G3, were identified. Our results indicate that this method may be useful for large-scale screens designed to identify novel radiation susceptibility genes.

Fine mapping of radiation susceptibility and gene expression analysis of LEC congenic rat lines.

LEC rats constitute an animal model of high susceptibility to X-rays. We developed congenic LEC rat lines (recipient strain, Fischer 344 (F344)) and performed genome-wide genotyping to identify radiation susceptibility genes. We mapped seven positional candidate genes, Bmp10, Gpr73, Gp9, Cnbp, Copg, Rab7, and Rpn1, to an approximately 1.2-Mb region located between loci D4Got85 and D4Got148 on chromosome 4. None of the seven genes has been reported to be associated with radiation susceptibility. Comparison of the coding sequences for these seven genes in F344 and LEC rats showed no changes in deduced amino acid sequences. We determined gene expression differences in Gp73, Gp9, and Cnbp as well as strain-specific variations in upstream sequences of these genes. Our results suggest that radiation susceptibility in the LEC rat is primarily attributable to one of the genes within this approximately 1.2-Mb region; however, expression analysis gave no clear indication as to which gene is responsible.

[RadGenomics project].

Human health conditions are largely determined by a complex interplay among genetic susceptibility, environmental factors, and aging. The RadGenomics project, which began in April 2001, promotes analysis of genes in response to irradiation, identification of their allelic variants in the human population, development of an effective procedure for quantitating individual radio-sensitivity, and analysis of the interrelationship between genetic heterogeneity and susceptibility to irradiation. Major groups of genes with which the project will concern itself include DNA repair genes, cell cycle genes, oncogenes, tumor suppressor genes, genes for programmed cell death, genes for signal transduction, and genes for oxidative processes. The outcome of the RadGenomics project should lead to improved protocols for personalized radiotherapy and reduce the possible side effects of treatment. The project will contribute to future research on the molecular mechanisms of radiation sensitivity in humans and stimulate the development of new high-throughput technology for a broader application of the biological and medical sciences. Identification of functionally important polymorphisms in the radiation response genes may determine individual differences in sensitivity to radiation exposure. The staff members, who are specialists in a variety of fields including genome science, radiation biology, medical science, molecular biology, and bioinformatics, have come to the RadGenomics project from various universities, companies, and research institutes.