HEG1 Protects Against Atherosclerosis by Regulating Stable Flow-Induced KLF2/4 Expression in Endothelial Cells

BACKGROUND

Atherosclerosis preferentially occurs in arterial regions of disturbed blood flow, and stable flow (s-flow) protects against atherosclerosis by incompletely understood mechanisms.

METHODS

Our single-cell RNA-sequencing data using the mouse partial carotid ligation model was reanalyzed, which identified Heart-of-glass 1 (HEG1) as an s-flow-induced gene. HEG1 expression was studied by immunostaining, quantitive polymerase chain reaction, hybridization chain reaction, and Western blot in mouse arteries, human aortic endothelial cells (HAECs), and human coronary arteries. A small interfering RNA-mediated knockdown of HEG1 was used to study its function and signaling mechanisms in HAECs under various flow conditions using a cone-and-plate shear device. We generated endothelial-targeted, tamoxifen-inducible HEG1 knockout (HEG1iECKO) mice. To determine the role of HEG1 in atherosclerosis, HEG1iECKO and littermate-control mice were injected with an adeno-associated virus-PCSK9 [proprotein convertase subtilisin/kexin type 9] and fed a Western diet to induce hypercholesterolemia either for 2 weeks with partial carotid ligation or 2 months without the surgery.

RESULTS

S-flow induced HEG1 expression at the mRNA and protein levels in vivo and in vitro. S-flow stimulated HEG1 protein translocation to the downstream side of HAECs and release into the media, followed by increased messenger RNA and protein expression. HEG1 knockdown prevented s-flow-induced endothelial responses, including monocyte adhesion, permeability, and migration. Mechanistically, HEG1 knockdown prevented s-flow-induced KLF2/4 (Kruppel-like factor 2/4) expression by regulating its intracellular binding partner KRIT1 (Krev interaction trapped protein 1) and the MEKK3-MEK5-ERK5-MEF2 pathway in HAECs. Compared with littermate controls, HEG1iECKO mice exposed to hypercholesterolemia for 2 weeks and partial carotid ligation developed advanced atherosclerotic plaques, featuring increased necrotic core area, thin-capped fibroatheroma, inflammation, and intraplaque hemorrhage. In a conventional Western diet model for 2 months, HEG1iECKO mice also showed an exacerbated atherosclerosis development in the arterial tree in both sexes and the aortic sinus in males but not in females. Moreover, endothelial HEG1 expression was reduced in human coronary arteries with advanced atherosclerotic plaques.

CONCLUSIONS

Our findings indicate that HEG1 is a novel mediator of atheroprotective endothelial responses to flow and a potential therapeutic target.

Single-cell and spatial profiling identify three response trajectories to pembrolizumab and radiation therapy in triple negative breast cancer

Strategies are needed to better identify patients that will benefit from immunotherapy alone or who may require additional therapies like chemotherapy or radiotherapy to overcome resistance. Here we employ single-cell transcriptomics and spatial proteomics to profile triple negative breast cancer biopsies taken at baseline, after one cycle of pembrolizumab, and after a second cycle of pembrolizumab given with radiotherapy. Non-responders lack immune infiltrate before and after therapy and exhibit minimal therapy-induced immune changes. Responding tumors form two groups that are distinguishable by a classifier prior to therapy, with one showing high major histocompatibility complex expression, evidence of tertiary lymphoid structures, and displaying anti-tumor immunity before treatment. The other responder group resembles non-responders at baseline and mounts a maximal immune response, characterized by cytotoxic T cell and antigen presenting myeloid cell interactions, only after combination therapy, which is mirrored in a murine model of triple negative breast cancer.

Characterization of CSF inflammatory markers after hemorrhagic stroke and their relationship to disease severity

Background

The inflammatory response within the central nervous system is a key driver of secondary brain injury after hemorrhagic stroke, both in patients with intracerebral hemorrhage (ICH) and aneurysmal subarachnoid hemorrhage (aSAH). In this study, we aimed to characterize inflammatory molecules in the blood and cerebrospinal fluid (CSF) of patients within 72 hours of hemorrhage to understand how such molecules vary across disease types and disease severity.

Methods

Biological samples were collected from patients admitted to a single-center Neurosciences Intensive Care Unit with a diagnosis of ICH or aSAH between 2014 and 2022. Control CSF samples were collected from patients undergoing CSF diversion for normal pressure hydrocephalus. A panel of immune molecules in the plasma and CSF samples was analyzed using Cytometric Bead Array assays. Clinical variables, including demographics, disease severity, and intensive care unit length of stay were collected.

Results

Plasma and/or CSF samples were collected from 260 patients (188 ICH patients, 54 aSAH patients, 18 controls). C-C motif chemokine ligand-2 (CCL2), interleukin-6 (IL-6), granulocyte-colony stimulating factor (G-CSF), interleukin-8 (IL-8), and vascular endothelial growth factor (VEGF), were detectable in the CSF within the first 3 days after hemorrhage, and all were elevated compared to plasma. Compared with controls, CCL2, IL-6, IL-8, G-CSF, and VEGF were elevated in the CSF of both ICH and aSAH patients (p<0.01 for all comparisons). VEGF was increased in ICH patients compared to aSAH patients (p<0.01). CCL2, G-CSF, and VEGF in the CSF were associated with more severe disease in aSAH patients only.

Conclusions

Within 3 days of hemorrhagic stroke, proinflammatory molecules can be detected in the CSF at higher concentrations than in the plasma. Early concentrations of some pro-inflammatory molecules may be associated with markers of disease severity.

OS10.6.A What is the initial cell in the subventricular zone for human glioblastoma genesis?

Background

We all have a fundamental question about why glioblastoma (GBM) develops. In order to find the answer to this fundamental question, if you find out what the first cell is, you will get closer to the answer. GBM arises from the subventricular zone (SVZ). GBM is one of the most devastating tumour of human brain as the most optimal treatment barely prolongs the survival, and it does not cure the disease. As the majority of GBM tissues show copy number variations (CNV) of co-altered chromosomal 7 gain and 10 loss, we hypothesized the origin cell (Oc) of SVZ may be traced back with these markers. The cellular identity of the Oc is still unknown and it is different from the tumour-derived progenitor-like cells. We aimed to define these cells from the SVZ that have a potential to get activated into GBM.

Material and Methods

We compared bulk RNA sequencing (RNAseq) data of IDHwt GBM tumor tissue (n=122), tumor free SVZ from GBM patients (n=40), tumor-free control SVZ of non-glial tumor (n=9). Pared single nucleus RNAseq (snRNAseq) or single cell RNAseq (scRNAseq) samples of tumor free SVZ (n=11) and GBM tumor (n=8), were done to see cell specific CNVs. We developed genetically engineered mouse models for GBM genesis introducing three driver mutations (TP53, PTEN, and EGFRviii) into SVZ to isolate mouse Oc (mOc) and mouse cancer cells (mCc). The biological characteristics of separated mOc and mCc were compared. Bulk RNAseq and scRNAseq were performed on these cells (mOc, mCc), and their cellular state was compared with the human gene set.

Results

In this work, we found two types of the Oc in the RNA sequencing of 60 human tumour free-SVZ samples. Furthermore, single-cell level analysis revealed that two Oc types in SVZ harbor ongoing patterns of CNV co-alterations from Oc1 to Oc2, and finally to GBM. The Oc1 type cells contained the CNV signature of Oc2 ancestor with neural progenitor cell (NPC) signature. Oc2 type cells expressed a high level of EGFR than other cells with astrocyte-like cell signature. Both of these cells expressed oligodendrocyte progenitor cell (OPC)-like signatures in the SVZ. We validated the human-based findings by using the P53/PTEN/EGFR-mutant mouse model with EGFR/tdTomato overexpression and P53/PTEN knockout in the SVZ cells. As a result, non-tumourigenic and highly motile Oc-like cell-states are found in the mouse models, supporting the firework-like migration pattern from the SVZ.

Conclusion

Our results demonstrate how members of Oc preoccupy the SVZ, known as the stem cell niche and give rise to the tumour. We anticipate that a new therapy may emerge by targeting the Oc in the SVZ.

Deep learning algorithm to predict pathologic complete response to neoadjuvant chemotherapy for breast cancer prior to treatment

600

Background: Pathologic complete response (PCR) to neoadjuvant chemotherapy (NAC) is associated with improved disease-free survival and overall survival in patients with breast cancer. Predicting PCR at the patient level prior to treatment initiation would allow physicians and patients to focus on therapies with the highest likelihood of success and minimize unnecessary toxicities from chemotherapy. We hypothesize that pre-treatment prediction of PCR is possible through a deep neural network algorithm trained on breast MRI imaging obtained prior to treatment. Methods: 126 tumors from patients treated with neoadjuvant chemotherapy for T3 stage breast cancer at a single institution from 2002 to 2006 were analyzed. In total, 3780 MRI slices were included. 3 MRI contrast phases (pre, immediate post, delayed post) from each slice were used as individual inputs to create separate model predictions of PCR. The 2-D CNN was trained over 50 epochs on the training set. The model was tested on the isolated test set (30% of samples). Results: Average model prediction accuracy over the total test set using a single phase of contrast (pre-, immediate post-, or delayed post-) was 90.4%. Concordance adjustment was conducted, with exclusion of slices that had produced discordant predictions across different contrast phase inputs. This resulted in an increase of overall model accuracy to 97.6%. Model accuracy was similar across subsets of age and tumor size despite differences in PCR rates. However, model accuracy was significantly lower in the triple-negative disease group. Conclusions: We demonstrate a deep neural network that accurately predicts PCR based on breast MRI imaging taken prior to NAC initiation. Our findings represent the promise of deep learning algorithms in providing personalized prognostic data for physicians and patients. [Table: see text]

Using Adversarial Images to Assess the Robustness of Deep Learning Models Trained on Diagnostic Images in Oncology

Deep learning (DL) models have rapidly become a popular and cost-effective tool for image classification within oncology. A major limitation of DL models is their vulnerability to adversarial images, manipulated input images designed to cause misclassifications by DL models. The purpose of the study is to investigate the robustness of DL models trained on diagnostic images using adversarial images and explore the utility of an iterative adversarial training approach to improve the robustness of DL models against adversarial images.

Exploring vulnerabilities of deep learning algorithms to adversarial images across oncologic imaging modalities.

Abstract 184: The utility of deep metric learning for breast cancer identification on mammographic images

Purpose: Although deep learning (DL) models have shown increasing ability to accurately classify diagnostic images in oncology, significantly large amounts of well-curated data are often needed to match human level performance. Given the relative paucity of imaging datasets for less prevalent cancer types, there is an increasing need for methods which can improve the performance of deep learning models trained using limited diagnostic images. Deep metric learning (DML) is a potential method which can improve accuracy in deep learning models trained on limited datasets. Leveraging a triplet-loss function, DML exponentially increases training data compared to a traditional DL model. In this study, we investigated the utility of DML to improve the accuracy of DL models trained to classify cancerous lesions found on screening mammograms.

Methods: Using a dataset of 2620 lesions found on routine screening mammogram, we trained both a traditional DL and DML models to classify suspicious lesions as cancerous or benign. The VGG16 architecture was used as the basis for the DL and DML models. Model performance was compared by calculating model accuracy, sensitivity, and specificity on a blinded test set of 378 lesions. In addition to individual model performance, we also measured agreement accuracy when both the DL and DML models were combined. Sub-analyses were conducted to identify phenotypes which were best suited for each model type. Both models underwent hyperparameters optimization to identify ideal batch size, learning rate, and regularization to prevent overfitting.

Results: We found that the combination of the traditional DL model with DML model resulted in the highest overall accuracy (78.7%) representing a 7.1% improvement compared to the traditional DL model (p&lt;.001). Alone, the traditional DL model had an improved accuracy compared to the DML model (71.4% vs 66.4%). The traditional DL model had a higher sensitivity (94.8% vs 73.6 %) , but lower specificity (34.7% vs 55.1%) compared the DML model. Sub-analyses suggested the traditional DL model was more accurate on higher density breasts, whereas the DML model was more accurate on lower density breasts. Additionally, the traditional DL model had the highest accuracy on oval shaped lesions, compared to the DML model which was most accurate on irregularly shaped breast lesions.

Conclusion: Our study suggests that addition of DML models with traditional DL models can improve diagnostic image classification performance in cancer. Our results suggest DML models may provide increased specificity and help with classification of unique populations often misclassified by traditional DL models. Further studied investigating the utility of DML on other cancer imaging tasks are necessary to successfully build more robust DL models in cancer imaging.

Citation Format: Justin Du, Sachin Umrao, Enoch Chang, Marina Joel, Aidan Gilson, Guneet Janda, Rachel Choi, Yongfeng Hui, Sanjay Aneja. The utility of deep metric learning for breast cancer identification on mammographic images [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 184.

Abstract PO-074: The impact of phenotypic bias in the generalizability of deep learning models in non-small cell lung cancer

Although deep learning analysis of diagnostic imaging has shown increasing effectiveness in modeling non-small cell lung cancer (NSCLC) outcomes, a minority of proposed deep learning algorithms have been externally validated. Given a majority of these models are built on single institutional datasets, their generalizability across the entire population remains understudied. Moreover, the effect of biases that exist among institutional training dataset on overall generalizability of deep learning prognostic models is unclear. We attempted to identify demographic and clinical characteristics which if over-represented within training data could affect the generalizability of deep learning models aimed at predicting survival in patients with non-small cell lung cancer (NSCLC). Using a dataset of pre-treatment CT images of 422 patients diagnosed with non-small cell lung cancer (NSCLC), we examined deep learning model performance across demographic and tumor specific factors. Demographic factors of interest included age and gender. Clinical factors of interest included tumor histology, overall stage, T-Stage, and N-Stage. The effect of bias among training data was examined by varying the representation of demographic and clinical populations within the training and validation datasets. Model generalizability was measured by comparing AUC values among validation datasets (biased versus unbiased). AUC was estimated using 1,000 bootstrapped samples of 400 patients from validation cohorts. We found training datasets with biased representation of NSCLC histologist to be associated with greatest decrease in generalizability. Specifically, we found over-representation of adenocarcinoma within training datasets to be associated with an AUC reduction of 0.320 (0.296 - 0.344 CI, p&lt;.001). Similarly over-representation of squamous cell carcinoma was associated with an AUC reduction of 0.177 (0.156 - 0.201 CI, p&lt;.001). Biases in age (AUC 0.103, p&lt;0.001), T stage (0.170, p=0.01 ), and N stage (0.120, p= 0.01) were also associated with reduced generalizability among deep learning models. Gender bias within training data was not associated with decreases in generalizability. Deep learning models of non-small cell lung cancer outcomes fail to generalize if trained on bias datasets. Specifically, overrepresentation of histologic subtypes may decrease the generalizability of deep learning models for NSCLC. Efforts to assure training data is representative of population demographics may lead to improved generalizability across more diverse patient populations.

Citation Format: Aidan Gilson, Justin Du, Guneet Janda, Sachin Umrao, Marina Joel, Rachel Choi, Roy Herbst, Harlan Krumholz, Sanjay Aneja. The impact of phenotypic bias in the generalizability of deep learning models in non-small cell lung cancer [abstract]. In: Proceedings of the AACR Virtual Special Conference on Artificial Intelligence, Diagnosis, and Imaging; 2021 Jan 13-14. Philadelphia (PA): AACR; Clin Cancer Res 2021;27(5_Suppl):Abstract nr PO-074.

Abstract PO-078: Exploring adversarial image attacks on deep learning models in oncology

Purpose: Deep learning (DL) models have shown the ability to automate the classification of diagnostic images used for cancer detection. Unfortunately, recent evidence has suggested DL models are also vulnerable to adversarial image attacks by manipulating image pixels to force models to make incorrect predictions with high confidence. The existence of adversarial images, which are imperceptible from unmodified images to the human eye, poses a roadblock to the safe implementation of DL models in clinical settings. The extent to which diagnostic imaging is vulnerable to adversarial image attacks remains underexplored. We investigated the effectiveness of adversarial imaging attacks on DL models for three common imaging tasks within oncology. Additionally, we explored whether adversarial image attack vulnerability could be used as a metric to improve deep learning model performance. Methods: We employed adversarial imaging attacks on DL models for three common imaging tasks within oncology: 1) Classifying malignant lung nodules on CT imaging, 2) Classifying brain metastases on MRI imaging, 3) Classifying malignant breast lesions on mammograms. To assess relative vulnerability to adversarial image attacks, we also employed two DL models on non-medical images: 1) CIFAR10, 2) MNIST. We considered three first-order adversarial attacks: Fast Gradient Sign Method, Projected Gradient Descent, and Basic Iterative Method. Vulnerability to adversarial image attacks was assessed by comparing model accuracy at fixed levels of image perturbations. Model performance was also measured after removing images which were most susceptible to adversarial imaging attacks. Results: We observed that all three diagnostic imaging types were susceptible to adversarial imaging attacks. Overall diagnostic images were more vulnerable to adversarial attacks compared to non-medical images. Mammograms [29.6% accuracy] appeared to be the most vulnerable to adversarial imaging attacks followed by lung CTs [30.6% accuracy] and brain MRIs [30.8% accuracy]. Finally, we determined that removing images most vulnerable to adversarial manipulation leads to improved deep learning model performance [Mammogram: 73.6 % accuracy, CT: 83.0% accuracy, MRI 84.2% accuracy]. Conclusion: Our study demonstrates that diagnostic imaging modalities in cancer are likely more vulnerable to adversarial attacks than non-medical images. Susceptibility to adversarial imaging attacks varies across different diagnostic imaging modalities. Adversarial susceptibility for an individual image can be used as a valuable metric to improve DL model performance on diagnostic images.

Citation Format: Marina Joel, Sachin Umrao, Enoch Chang, Rachel Choi, Daniel Yang, Aidan Gilson, Roy Herbst, Harlan Krumholz, Sanjay Aneja. Exploring adversarial image attacks on deep learning models in oncology [abstract]. In: Proceedings of the AACR Virtual Special Conference on Artificial Intelligence, Diagnosis, and Imaging; 2021 Jan 13-14. Philadelphia (PA): AACR; Clin Cancer Res 2021;27(5_Suppl):Abstract nr PO-078.

Cellular, matrix, and mechano-biological differences in load-bearing versus positional tendons throughout development and aging: a narrative review

PURPOSE

Summarise available evidence comparing the cellular, biochemical, structural and biomechanical properties, and the changes that occur in these parameters in response to stimuli, in differentially loaded tendons across different stages of life.

METHODS

The PubMed database was searched for literature pertaining to differences between tendons using the term "tendon" or "tendinopathy", plus one or more of the following descriptors: "loading", "positional", "weight- or load-bearing", and "energy-storing". The abstracts were reviewed and relevant full-length articles retrieved and used to assemble a narrative review.

RESULTS

The incidence and prevalence of tendon disorders ("tendinopathies") is increasing in Western societies, with limited evidence that currently available treatments have any significant long-term effect on the disease course. A key emerging hypothesis is that disease in different tendons and even different regions within a tendon may be distinct. The available literature indicates that there are phenotypic differences, not only in the constitutive compositional and material properties but also in resident cells of positional compared with load-bearing tendons. Evident during early tendon growth, such differences have become well established by adulthood.

CONCLUSIONS

The pheno-endotype of tendinopathy may be distinct between load-bearing tendons compared to positional tendons, which has translational implications with regard to preventing and managing tendinopathy. Better understanding of the molecular, cellular, and biomechanical pathophysiology underlying disease phenotypes, will allow more targeted/personalised treatment and therefore improve outcomes.

Proteomic Characterization of Middle Ear Fluid Confirms Neutrophil Extracellular Traps as a Predominant Innate Immune Response in Chronic Otitis Media

BACKGROUND

Chronic Otitis Media (COM) is characterized by middle ear effusion (MEE) and conductive hearing loss. MEE reflect mucus hypersecretion, but global proteomic profiling of the mucosal components are limited.

OBJECTIVE

This study aimed at characterizing the proteome of MEEs from children with COM with the goal of elucidating important innate immune responses.

METHOD

MEEs were collected from children (n = 49) with COM undergoing myringotomy. Mass spectrometry was employed for proteomic profiling in nine samples. Independent samples were further analyzed by cytokine multiplex assay, immunoblotting, neutrophil elastase activity, next generation DNA sequencing, and/or immunofluorescence analysis.

RESULTS

109 unique and common proteins were identified by MS. A majority were innate immune molecules, along with typically intracellular proteins such as histones and actin. 19.5% percent of all mapped peptide counts were from proteins known to be released by neutrophils. Immunofluorescence and immunoblotting demonstrated the presence of neutrophil extracellular traps (NETs) in every MEE, along with MUC5B colocalization. DNA found in effusions revealed unfragmented DNA of human origin.

CONCLUSION

Proteomic analysis of MEEs revealed a predominantly neutrophilic innate mucosal response in which MUC5B is associated with NET DNA. NETs are a primary macromolecular constituent of human COM middle ear effusions.

The supination assessment task: An automated method for quantifying forelimb rotational function in rats

BACKGROUND

Neurological injuries or disease can impair the function of motor circuitry controlling forearm supination, and recovery is often limited. Preclinical animal models are essential tools for developing therapeutic interventions to improve motor function after neurological damage. Here we describe the supination assessment task, an automated measure of quantifying forelimb supination in the rat.

NEW METHOD

Animals were trained to reach out of a slot in a cage, grasp a spherical manipulandum, and supinate the forelimb. The angle of the manipulandum was measured using a rotary encoder. If the animal exceeded the predetermined turn angle, a reward pellet was delivered. This automated task provides a large, high-resolution dataset of turn angle over time. Multiple parameters can be measured including success rate, peak turn angle, turn velocity, area under the curve, and number of rotations per trial. The task provides a high degree of flexibility to the user, with both software and hardware parameters capable of being adjusted.

RESULTS

We demonstrate the supination assessment task can effectively measure significant deficits in multiple parameters of rotational motor function for multiple weeks in two models of ischemic stroke.

COMPARISON WITH EXISTING METHODS

Preexisting motor assays designed to measure forelimb supination in the rat require high-speed video analysis techniques. This operant task provides a high-resolution, quantitative end-point dataset of turn angle, which obviates the necessity of video analysis.

CONCLUSIONS

The supination assessment task represents a novel, efficient method of evaluating forelimb rotation and may help decrease the cost and time of running experiments.

Canine pancreatic-specific lipase concentrations in dogs with heart failure and chronic mitral valvular insufficiency

BACKGROUND

Chronic mitral valvular insufficiency (CMVI) in dogs is very common and might cause clinical signs of congestion and poor tissue perfusion.

HYPOTHESIS

Poor tissue perfusion from CMVI causes pancreatitis in dogs, as indicated by serum pancreatic lipase concentrations.

ANIMALS

Sixty-two client-owned dogs consisting of 40 dogs with different stages of heart failure from CMVI and 22 age-matched healthy dogs, based on full cardiac exam and routine laboratory tests.

METHODS

Prospective, controlled, observational study. Serum canine pancreatic lipase immunoreactivity (cPLI) concentrations were determined by quantitative cPLI test in healthy and CMVI groups.

RESULTS

Serum cPLI concentrations were 54.0 μg/L (IQR: 38.0-78.8 μg/L) in control, 55.0 μg/L (IQR: 38.3-88.8 μg/L) in ISACHC I, 115.0 μg/L (IQR: 45.0-179.0 μg/L) in ISACHC II and 223.0 μg/L (IQR: 119.5-817.5 μg/L) in ISACHC III. Close correlation to serum cPLI concentration was found in the left atrial to aorta (LA/Ao) ratio (r = 0.597; P = .000) and the severity of heart failure (r = 0.530; P = .000).

CONCLUSIONS AND CLINICAL IMPORTANCE

This study found CMVI is associated with pancreatic injury in congestive heart failure caused by CMVI. Therefore, periodic monitoring on cPLI could be useful in monitoring dogs in heart failure.

Comparison of 2 retrieval devices for heartworm removal in 52 dogs with heavy worm burden

BACKGROUND

For treating dogs with heavy heartworm infection, mechanical removal using various retrieval devices is useful. However, the efficacy and safety of retrieval devices have rarely been studied.

HYPOTHESIS

Catheter-based heartworm removal using 2 retrieval devices (basket and tripod grasping forceps) is efficient and safe for treating dogs with heavy worm burden.

ANIMALS

Fifty-two client-owned dogs with heavy (Class III and IV) worm burden.

METHODS

A retrospective study was performed on 52 dogs, using a catheter-based heartworm removal approach using 2 types of retrieval devices (ie, the basket and the tripod grasping forceps). The efficacy and complications associated with the 2 devices were assessed.

RESULTS

The basket device was used on 22 of the study group dogs, and the tripod grasping forceps was used on 30 of the dogs. The postoperative survival rate was 95.5% for the basket device and 80% for the tripod grasping forceps, but the difference was not statistically significant. The worm number captured per attempt was 3.5 ± 1.7 using the basket device and 1.9 ± 0.85 for the tripod grasping forceps (P < .05). Various complications associated with heartworm removal were noticed with both retrieval devices.

CONCLUSIONS AND CLINICAL IMPORTANCE

This study suggests that catheter-based heartworm removal is not only a relatively safe and efficient therapeutic method in dogs with heavy worm burden, but more efficient using the basket device. Our data do not indicate a clear safety advantage between the 2 devices evaluated, although the survival rate was numerically higher in dogs undergoing a basket intervention.

Effects of Structural Properties of Hf-Based Gate Stack on Transistor Performance

Electron traps in ALD and MOCVD HfO2 and HfSiO high-k dielectrics were investigated using both conventional DC and pulse measurements. It was found that the traps in the gate stack could be associated with defects of different activation energies and capture cross-sections. This points to potentially different origins of the electrically active defects, which can be either intrinsic or process-related. Structural non-uniformity of the high-k film, associated with grain formation and phase separation, may lead to variation of electrical properties of the gate dielectric along the transistor channel. Effects of such dielectric non-uniformity, as well as electron trapping, on the measured transistor mobility were evaluated.

Comparison of the pathogenicity in three different Korean canine parvovirus 2 (CPV-2) isolates

Canine parvovirus type 2 (CPV-2) is a major pathogen inducing acute hemorrhagic gastroenteritis in dogs. Despite the identification of numerous CPV-2 variants (from CPV-2a to CPV-2c), the pathogenic differences among the CPV-2 variants in dogs have not been evaluated. The aim of this study was to compare the pathogenicity of CPV-2 variants (CPV-2a-I, CPV-2a-V and CPV-2b) isolated mainly from Korea. We evaluated the pathogenicity of three different CPV-2 variants, by performing clinical, hematological, serological and histopathological examinations after experimentally inoculating three types of CPV-2 variants into young puppies. We found that the overall pathogenicity of the CPV-2a variants (CPV-2a-I and 2a-V) was severer compared to the CPV-2b variant. In addition, there was no significant difference in pathogenicity between the two CPV-2a variants. Our findings indicate that there are differences in the pathogenicity of CPV-2 variants in dogs, which may be useful to understand the different pathobiology of the CPV-2 variants.

HIV type 1 Gag and nucleocapsid proteins: cytoskeletal localization and effects on cell motility

Cell motility is likely to play a pivotal role in HIV infection by promoting the dissemination of infected cells. On the basis of observations indicating an interaction between HIV-1 Gag and target cell filamentous actin, we hypothesized that these interactions would promote cell motility of HIV-infected cells. Indeed, we have found that HIV-1 infection enhances the chemotactic response of macrophages. To specifically investigate the significance of the interactions between Gag and cellular actin, we transfected NIH 3T3 fibroblasts and HeLa cells with a construct that permits the expression of HIV-1 Gag in the absence of any other viral protein. Fractionation experiments showed that Gag was present in cytoskeletal fraction containing long actin filaments and in a high-speed postcytoskeletal fraction with short actin filaments. We have also localized HIV-1 Gag to the lamellipodia of chemoattractant-stimulated cells. Significantly, the motility of Gag-expressing cells was enhanced in chemotaxis assays. In vitro mutagenesis experiments showed that HIV-1 Gag binds filamentous actin through the nucleocapsid domain (NC). An NC-green fluorescent protein fusion had the same cellular distribution as the complete protein, and its expression increased cell motility. These data suggest that interactions between HIV-1 Gag and actin in infected cells enhance cell motility. Ultimately this enhanced motility of infected cells could promote the dissemination of virus into the brain and other tissues.

Rotatory subluxation of the scaphoid

Rotatory subluxation of the scaphoid remains a challenging problem; a thorough understanding of the basic anatomy, particularly the ligament complexes of the wrist, as well as wrist kinematics are required for the proper diagnosis and treatment. Our understanding of rotatory subluxation of the scaphoid has improved over the years and continues to expand.

The wound healing response after laser in situ keratomileusis and photorefractive keratectomy: elusive control of biological variability and effect on custom laser vision correction

Biological diversity in the wound healing response is thought to be a major factor limiting the predictability of the outcome of refractive surgical procedures such as laser in situ keratomileusis and photorefractive keratectomy. Corneal wound healing is critical to the success of topography-linked or wave front-linked excimer laser ablation to optimize visual performance. This is because of the importance of retaining subtle features of custom ablation and the tendency of epithelial hyperplasia and stromal remodeling to obscure these features following either procedure. The corneal wound healing response is exceedingly complex. Keratocyte apoptosis, which occurs in response to epithelial injury, is the earliest observable event in the wound healing cascades and is therefore an excellent target for pharmacological intervention. Alterations of surgical technique can be designed to limit keratocyte apoptosis and the subsequent events in corneal wound healing. Abnormalities of the cascades could contribute to the pathogenesis of corneal diseases. For example, recent data have suggested that perturbation of the keratocyte apoptosis/mitosis balance could underlie the development of keratoconus in a proportion of patients.

IL-4 and IL-10 are both required for the induction of oral tolerance

Protection from the development of experimental autoimmune uveitis (EAU) can be induced by feeding mice interphotoreceptor retinoid binding protein before uveitogenic challenge with the same protein. Two different regimens are equally effective in inducing protective tolerance, although they seem to do so through different mechanisms: one involving regulatory cytokines (IL-4, IL-10, and TGF-beta), and the other with minimal involvement of cytokines. Here we studied the importance of IL-4 and IL-10 for the development of oral tolerance using mice genetically engineered to lack either one or both of these cytokines. In these animals we were able to protect against EAU only through the regimen inducing cytokine-independent tolerance. When these animals were fed a regimen that in the wild-type animal is thought to predominantly induce regulatory cells and is associated with cytokine secretion, they were not protected from EAU. Interestingly, both regimens were associated with reduced IL-2 production and proliferation in response to interphotoreceptor retinoid binding protein. These findings indicate that both IL-4 and IL-10 are required for induction of protective oral tolerance dependent on regulatory cytokines, and that one cytokine cannot substitute for the other in this process. These data also underscore the fact that oral tolerance, manifested as suppression of proliferation and IL-2 production, is not synonymous with protection from disease.

Successful anastomosis between tissue-engineered intestine and native small bowel

BACKGROUND

Previous work from this laboratory has shown that isolated intestinal epithelial organoid units on porous biodegradable polymer scaffolds formed vascularized cysts lined by a neomucosa. The purpose of this study was to demonstrate anastomosis between tissue-engineered intestine and the native small bowel and to observe the effect of this anastomosis on cyst growth.

METHODS

Intestinal epithelial organoid units from neonatal Lewis rats were seeded onto porous biodegradable polymer tubes made of polyglycolic acid, and they were implanted into the omentum of adult male Lewis rats. Three weeks after implantation, the unit-polymer constructs were anastomosed in a side-to-side fashion to the native jejunum in 20 rats (group 1). The other 18 rats were closed without anastomosis (group 2). All 38 tissue-engineered constructs were harvested 10 weeks after implantation. Four rats underwent upper gastrointestinal (GI) study before they were killed.

RESULTS

The rats in group 1 increased their body weights equal to those in group 2, and there was no statistically significant difference between the two groups. Upper GI examinations revealed no evidence of either bowel stenosis or obstruction at the anastomotic site. Grossly, the patency of the anastomosis was 90% and the lumen of the cyst was visualized by the upper GI study. At the second operation, there was no significant difference in the size of the cysts in either group: however, at the time the rats were killed, the length of the cysts in group 1 was significantly longer than that in group 2 (P<0.05 using Mann-Whitney U test). Histological examination showed that cysts after anastomosis were lined by a neomucosa in continuity to native small bowel across the anastomotic site and also demonstrated crypt-villus structures. Morphometric study demonstrated that cysts in group 1 had significantly greater villus number, height, and surface length than did those in group 2.

CONCLUSIONS

Anastomosis between tissue-engineered intestine and native small bowel resulted in no complications after the operation, kept a high patency rate, and maintained mucosal continuity between the tissue-engineered intestine and native small bowel. Furthermore, anastomosis had a positive effect on cyst size and development of the mucosa in the tissue-engineered intestine.

Ex vivo expansion and differentiation of unselected peripheral blood progenitor cells in serum-free media

The ability to expand and differentiate unselected PBPC was investigated. Cells were grown in serum-free media containing stem cell factor, GCSF and megakaryocyte growth and development factor (pegylated PEG-rHuMGDF) with or without supplemental serum. Optimal proliferation occurred when PBPC were cultured without prior Ficoll-Paque separation in serum-free media. Cell yields after 17 days of culture were proportional to the percentage of CD34+ cells in the starting population and were 1170+/-302-fold higher than the starting numbers of CD34+ cells. Granulocyte-macrophage colony-forming units increased over 12 days of culture, whereas the numbers of erythroid colony-forming cells peaked between 4 and 7 days. Elimination of PEG-rHuMGDF from cell cultures resulted in significantly lower yields of myeloid and erythroid colony-forming cells and total cell numbers. Cell differentiation into neutrophils was indicated by progressive increases in CD11b, CD15, and CD66b expression. Expanded neutrophils phagocytosed and killed bacteria as efficiently as neutrophils from normal donors. Large-scale expansion studies yielded similar proliferation and differentiation results as parallel small-scale cultures. Therefore, unselected PBPC can be efficiently expanded and differentiated into large numbers of functional mature neutrophils.

Biological and biochemical modulation of Trichomonas vaginalis KT9 isolate after shifting of culture medium from TPS-1 into TYM

To evaluate the biological and biochemical characteristics of Trichomonas vaginalis KT9 isolate, the growth and size of trichomonads, pathogenicity in mouse, protein profiles and proteinase activity were examined after shifting the medium from TPS-1 into TYM. Generation time of trichomonads in TYM medium was 4.5 hr in comparison to TPS-1 with 7.1 hr. Size of trichomonads cultured in TPS-1 medium (8.5 +/- 0.9 x 6.0 +/- 0.9 microns) was significantly smaller than those in TYM medium (10.9 +/- 1.4 x 8.2 +/- 0.9 microns). Trichomonads cultured in TYM medium produced subcutaneous abscess in 9 out of 10 mice, whereas those in TPS-1 medium produced abscesses in 2 out of 10 mice. In SDS-PAGE, trichomonad lysates from both media showed ten common bands. However, trichomonads in TYM medium showed additional bands of 136 kDa, 116 kDa and 40 kDa in comparison to those in TPS-1 with 100 kDa. By immunoblot with T. vaginalis-immunized rabbit sera, T. vaginalis cultivated in both TYM and TPS-1 media showed 5 common bands, and unique bands of 116 kDa, 105 kDa, and 86 kDa were observed in trichomonads in TYM while a 140 kDa band in those in TPS-1. In gelatin SDS-PAGE, trichomonads in TYM degraded gelatin stronger than those in TPS-1. Also protease activity of trichomonads in TYM was significantly higher than that of trichomonads in TPS-1 using Bz-Pro-Phe-Arg-Nan as a substrate. According to the results, it is assumed that the shift from TPS-1 into TYM medium for cultivation of T. vaginalis might modulate the biological and biochemical properties of T. vaginalis in vitro.

Serum complement determinations in patients with quiescent systemic lupus erythematosus

OBJECTIVE

To determine whether complement component analyses during a period of inactive disease can define clinically important subgroups and predict morbidity in patients with systemic lupus erythematosus (SLE).

METHODS

We identified 277 patients with SLE whose disease became clinically inactive at some point after diagnosis. Serum samples were obtained at that time and tested for total complement activity (CH100) and antigenic levels of C1q, C1r, C1s, C3 and C4. Results of complement determinations were correlated with demographic characteristics and clinical findings in the followup period (mean observation period 4.25 years).

RESULTS

We identified 25 (9%) patients with multiple complement determinations below the normal range. 24 other patients (8.5%) had a very low level of a single complement component. The group with multiple complement determinations below the normal range was much more likely than the normocomplementemic SLE controls to progress to renal insufficiency. In other respects, complement component determinations were neither reflective nor predictive of clinical course.

CONCLUSION

In this group of patients with inactive SLE, complement component analyses did not generally correlate with longterm outcome; however, multiple low complement component determinations during disease quiescence was associated with increased risk of renal insufficiency.

Anaesthetic management of caesarean section in a patient with active recurrent genital herpes and AIDS-related dementia

We report the anaesthetic management of a pregnant patient with multiple manifestations of HIV infection who underwent Caesarean section. A 30-yr-old, HIV-positive, Haitian woman presented with acute psychosis at 28 weeks' gestation. A diagnosis of HIV dementia complex was made and haloperidol therapy was started. Five days after admission the patient was found to be in labour and tocolytic therapy with terbutaline was commenced. A vaginal lesion compatible with herpes simplex virus was observed which was treated with acyclovir. After 3 days of tocolytic therapy there were no further signs of preterm labour. Two weeks later, at 30 weeks' gestation, the patient's membranes ruptured spontaneously. The herpes labialis lesion was still present and urgent Caesarean section was begun using subarachnoid 0.75% bupivacaine 1.5 ml. The patient had no intraoperative problems and a 1700-g healthy male child was delivered.

Female surgeons in the 1990s. Academic role models

As part of a larger survey of the membership of the Association of Women Surgeons, data were collected to characterize the participation of women in academic surgery and surgical education. Of the 1500 members, 676 (45%) responded, and 318 of these respondents held a faculty appointment. Two hundred thirty-nine of the respondents' practices were based in a university or university-affiliated hospital. This sample was biased toward more academically motivated, upwardly mobile surgeons. Two hundred thirty-six of the respondents believed policies relating to tenure are unfair to women. Four hundred fifty-three respondents reported having role models at some point in their career, half of whom were men. However, only 204 still had access to role models or mentors at the time of the survey. Six hundred thirty respondents believed that female medical students need successful female surgeons as role models.

[Blastogenesis of splenic lymphocytes to specific antigens and PHA in Paragonimus westermani infected mice]

Paragonimus westermani is a common fluke in Korea. The present study aimed to observe the cell mediated immune response in experimental paragonimiasis of mice. The mouse (BALB/c) was orally inoculated with 40 metacercariae of P. westermani from Cambaroides similis. During the infection (1, 2, 4, 6 weeks) of mouse, blastogenic response of splenic lymphocytes to P. westermani adult antigen, metacercaria antigen, and PHA were observed. Sera from infected and noninfected mice added to normal mouse splenic lymphocytes with or without PHA. The blastogenic response of splenic lymphocytes to PHA was reduced after 1 week of infection. However after 6 weeks of infection, the response was restored to the control level. The blastogenic response of splenic lymphocytes to P. westermani adult or metacercaria antigen increased significantly on 1 week after infection, and maintained up to 6 weeks after infection. The response of non-infected mice was suppressed by addition of the infected mouse serum. The present results suggested that cellular immunity was involved in P. westermani infected mice and that P. westermani anti-serum inhibited proliferation of T lymphocytes.