Antimicrobial Activity of Amino-Derivatized Cationic Polysaccharides.

To improve the antimicrobial property of chitosan, water-soluble chitosan modified in their quaternary ammonium groups were synthesized. The antimicrobial properties were evaluated against Escherichia coli, Bacillus subtilis, Saccharomyces cerevisiae and Candida tropicalis. The activities increased with increasing cationic charges and the length of the alkyl chain as follows amino-chitosan, dimethylaminoethyl-chitosan, dimethylpropyl amino-chitosan, dimethylamino-1-propyl-chitosan, diethylaminoethyl (DEAE)-chitosan, and quaternized DEAE-chitosan. The modified cationic chitosans showed high antimicrobial property against B. subtilis-Gram-positive bacteria, but were less active towards yeast (C. tropicalis and S. cerevisiae) and E. coli-Gram-negative bacteria. The simple structure of the Gram-positive bacteria may explain why the cationic chitosan derivatives are more active towards B. subtilis than yeast and E. coli. The target sites of the chitosan derivatives are assumed to be the cytoplasmic membranes of microorganisms. The antimicrobial activities were strongly dependent on the cationic charge and the molecular weight. It can be suggested that these cationic chitosan derivatives have potential as antimicrobial agents.

Deep neural network with weight sparsity control and pre-training extracts hierarchical features and enhances classification performance: Evidence from whole-brain resting-state functional connectivity patterns of schizophrenia.

Functional connectivity (FC) patterns obtained from resting-state functional magnetic resonance imaging data are commonly employed to study neuropsychiatric conditions by using pattern classifiers such as the support vector machine (SVM). Meanwhile, a deep neural network (DNN) with multiple hidden layers has shown its ability to systematically extract lower-to-higher level information of image and speech data from lower-to-higher hidden layers, markedly enhancing classification accuracy. The objective of this study was to adopt the DNN for whole-brain resting-state FC pattern classification of schizophrenia (SZ) patients vs. healthy controls (HCs) and identification of aberrant FC patterns associated with SZ. We hypothesized that the lower-to-higher level features learned via the DNN would significantly enhance the classification accuracy, and proposed an adaptive learning algorithm to explicitly control the weight sparsity in each hidden layer via L1-norm regularization. Furthermore, the weights were initialized via stacked autoencoder based pre-training to further improve the classification performance. Classification accuracy was systematically evaluated as a function of (1) the number of hidden layers/nodes, (2) the use of L1-norm regularization, (3) the use of the pre-training, (4) the use of framewise displacement (FD) removal, and (5) the use of anatomical/functional parcellation. Using FC patterns from anatomically parcellated regions without FD removal, an error rate of 14.2% was achieved by employing three hidden layers and 50 hidden nodes with both L1-norm regularization and pre-training, which was substantially lower than the error rate from the SVM (22.3%). Moreover, the trained DNN weights (i.e., the learned features) were found to represent the hierarchical organization of aberrant FC patterns in SZ compared with HC. Specifically, pairs of nodes extracted from the lower hidden layer represented sparse FC patterns implicated in SZ, which was quantified by using kurtosis/modularity measures and features from the higher hidden layer showed holistic/global FC patterns differentiating SZ from HC. Our proposed schemes and reported findings attained by using the DNN classifier and whole-brain FC data suggest that such approaches show improved ability to learn hidden patterns in brain imaging data, which may be useful for developing diagnostic tools for SZ and other neuropsychiatric disorders and identifying associated aberrant FC patterns.

A predictive model for lymph node metastasis using tumor location in presumed early-stage endometrioid endometrial cancer patients.

OBJECTIVE: The aim of this study was to identify high- and low-risk subgroups of patients with lymph node (LN) metastasis in presumed early-stage endometrioid endometrial cancer (EC) patients. METHODS: Clinicopathologic data of presumed early-stage endometrioid EC patients (n=361) treated with lymphadenectomy between March 2000 and July 2022 were analyzed. None of the patient had definite evidence of LN metastasis in a preoperative magnetic resonance imaging (MRI). A received operating characteristic curve analysis was conducted to define the sensitivity and specificity for the combined preoperative risk factors for LN metastasis, which was determined by multivariate analysis. RESULTS: Nineteen patients (5.3%) had LN metastasis. Multivariate analysis identified cervical stromal invasion on MRI (odds ratio [OR]=4.386; 95% confidence interval [CI]=1.020-18.852; p=0.047), cornual location of tumor on MRI (OR=36.208; 95% CI=7.902-165.913; p<0.001), and lower uterine segment/isthmic location of tumor on MRI (OR=8.454; 95% CI=1.567-45.610; p=0.013) as independent prognostic factors associated with LN metastasis. Patients were categorized into low- and high-risk groups according to risk criteria. Significant differences in the rates of LN metastasis were observed between the two groups (0.4% vs. 22.2%, p<0.001). CONCLUSION: Approximately 95% of presumed early-stage endometrioid EC patients did not have LN metastasis. A model using tumor location was significantly correlated with the risk of LN metastasis. Even in presumed early-stage endometrioid EC patients, therefore, tumor location should be investigated to determine whether to perform LN assessment.

Enhancement of xylitol production in glycerol kinase disrupted Candida tropicalis by co-expression of three genes involved in glycerol metabolic pathway.

Glycerol can be used as a primary carbon source by yeasts, little is known regarding glycerol metabolism in Candida tropicalis. In this study, glycerol kinase gene (gk) was disrupted from xylitol dehydrogenase gene (XYL2) knockout C. tropicalis strain BSXDH-3. The resultant gk knockout C. tropicalis strain was incapable to grow on glycerol. The cells growth on glycerol was resumed by co-expressing Scheffersomyces stipitis gcy1, 2 and 3 genes, which respectively encode NADP(+)-dependent glycerol dehydrogenase 1, 2 and 3, under the control of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) promoter. NADPH-dependent xylitol production was higher in the engineered strain, termed "GK", than in BSXDH-3. In fermentation experiments using glycerol as co-substrate with xylose, strain GK produced xylitol 0.85 and 1.28 g l(-1) h(-1) at the time periods of 16 and 24 h, respectively, which is 30 and 18 % higher at same time intervals in BSXDH-3. This is the first report of gk gene disruption and co-expression of gcy1, 2 and 3 genes for NADPH regeneration and enhanced xylitol production in C. tropicalis.

Iterative approach of dual regression with a sparse prior enhances the performance of independent component analysis for group functional magnetic resonance imaging (fMRI) data.

This study proposes an iterative dual-regression (DR) approach with sparse prior regularization to better estimate an individual's neuronal activation using the results of an independent component analysis (ICA) method applied to a temporally concatenated group of functional magnetic resonance imaging (fMRI) data (i.e., Tc-GICA method). An ordinary DR approach estimates the spatial patterns (SPs) of neuronal activation and corresponding time courses (TCs) specific to each individual's fMRI data with two steps involving least-squares (LS) solutions. Our proposed approach employs iterative LS solutions to refine both the individual SPs and TCs with an additional a priori assumption of sparseness in the SPs (i.e., minimally overlapping SPs) based on L(1)-norm minimization. To quantitatively evaluate the performance of this approach, semi-artificial fMRI data were created from resting-state fMRI data with the following considerations: (1) an artificially designed spatial layout of neuronal activation patterns with varying overlap sizes across subjects and (2) a BOLD time series (TS) with variable parameters such as onset time, duration, and maximum BOLD levels. To systematically control the spatial layout variability of neuronal activation patterns across the "subjects" (n=12), the degree of spatial overlap across all subjects was varied from a minimum of 1 voxel (i.e., 0.5-voxel cubic radius) to a maximum of 81 voxels (i.e., 2.5-voxel radius) across the task-related SPs with a size of 100 voxels for both the block-based and event-related task paradigms. In addition, several levels of maximum percentage BOLD intensity (i.e., 0.5, 1.0, 2.0, and 3.0%) were used for each degree of spatial overlap size. From the results, the estimated individual SPs of neuronal activation obtained from the proposed iterative DR approach with a sparse prior showed an enhanced true positive rate and reduced false positive rate compared to the ordinary DR approach. The estimated TCs of the task-related SPs from our proposed approach showed greater temporal correlation coefficients with a reference hemodynamic response function than those of the ordinary DR approach. Moreover, the efficacy of the proposed DR approach was also successfully demonstrated by the results of real fMRI data acquired from left-/right-hand clenching tasks in both block-based and event-related task paradigms.

Factors associated with medical cost among patients with terminal cancer in hospice units.

This study identified factors associated with higher medical costs for patients with terminal cancer in hospice units in order to develop a daily payment system for hospice services within Korea's National Health Insurance (NHI) program. Through chart reviews conducted by staff nurses, medical information and costs were obtained for 274 patients with terminal cancer in 20 hospice units in October 2007. The daily medical cost per patient was calculated based on the fee-for-service scheme. The characteristics of the hospice units were examined by means of a semistructured questionnaire administered to hospice unit coordinators. Higher daily costs were associated with general hospital-based hospice units (as compared with free-standing units: p<0.01), low Palliative Performance Scale scores (PPS<50, p<0.05), and the presence of fever (p<0.01). In multivariate analysis, hospice unit type was found to be the factor most strongly associated with medical cost. A hospice payment system based on patient characteristics should be thoroughly considered.

Enhancement of xylitol production in Candida tropicalis by co-expression of two genes involved in pentose phosphate pathway.

The yeast Candida tropicalis produces xylitol, a natural, low-calorie sweetener whose metabolism does not require insulin, by catalytic activity of NADPH-dependent xylose reductase. The oxidative pentose phosphate pathway (PPP) is a major basis for NADPH biosynthesis in C. tropicalis. In order to increase xylitol production rate, xylitol dehydrogenase gene (XYL2)disrupted C. tropicalis strain BSXDH-3 was engineered to co-express zwf and gnd genes which, respectively encodes glucose-6-phosphate dehydrogenase (G6PDH) and 6-phosphogluconate dehydrogenase (6-PGDH), under the control of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) promoter. NADPH-dependent xylitol production was higher in the engineered strain, termed "PP", than in BSXDH-3. In fermentation experiments using glycerol as a co-substrate with xylose, strain PP showed volumetric xylitol productivity of 1.25 g l(-1) h(-1), 21% higher than the rate (1.04 g l(-1) h(-1)) in BSXDH-3. This is the first report of increased metabolic flux toward PPP in C. tropicalis for NADPH regeneration and enhanced xylitol production.

N-glycosylation status of beta-haptoglobin in sera of patients with colon cancer, chronic inflammatory diseases and normal subjects.

N-glycosylation status of purified beta-haptoglobin from sera of 17 patients, and from sera of 14 healthy volunteer subjects, was compared by blotting with various lectins and antibodies. Patients in this study were diagnosed as having colon cancer through histological examination of each tumor tissue by biopsy. Blotting index of serum beta-haptoglobin with Aleuria aurantia lectin (AAL) was clearly higher for cancer patients than for healthy subjects. No such distinction was observed for blotting with three other lectins and two monoclonal antibodies. To determine tumor-associated reactivity of AAL binding as compared to inflammatory processes in colonic tissues, beta-haptoglobin separated from sera of 5 patients with Crohn's disease (CD), and 4 patients with ulcerative colitis (UC), was studied. All these cases, except one case of UC, showed AAL index lower than that in cancer cases, similarly to healthy subjects. The higher AAL binding of beta-haptoglobin in colon cancer patients than in healthy subjects appeared to be due to alpha-L-fucosyl residue, since it was eliminated by bovine kidney alpha-fucosidase treatment. N-linked glycans of serum haptoglobin from colon cancer patients vs. healthy subjects were released by N-glycanase, fluorescence-labeled, and subjected to normal-phase high performance liquid chromatography (NP-HPLC). Glycan structures were determined based on glucose unit (GU) values and their changes upon sequential treatment with various exoglycosidases. Glycosyl sequences and their branching status of glycans from 14 cases of serum beta-haptoglobin were characterized. The identified glycans were sialylated or nonsialylated, bi-antennary or tri-antennary structures, with or without terminal fucosylation.

Antitumor activity of methylan polysaccharide derivatives.

Methylan polysaccharide derivatives were prepared by dialkylaminoalkylation and reductive amination followed by quaternization. Their antitumor activity was investigated and a relationship between structure and activity is suggested. For quaternized DEAE-methylan at only 75 mug ml(-1), tumor cell proliferation was suppressed by 58-84% in three cell lines tested in the order Colo < Hela < HepG2.

Sequential and simultaneous statistical optimization by dynamic design of experiment for peptide overexpression in recombinant Escherichia coli.

The production of recombinant anti-HIV peptide, T-20, in Escherichia coli was optimized by statistical experimental designs (successive designs with multifactors) such as 2(4-1) fractional factorial, 2(3) full factorial, and 2(2) rotational central composite design in order. The effects of media compositions (glucose, NPK sources, MgSO4, and trace elements), induction level, induction timing (optical density at induction process), and induction duration (culture time after induction) on T-20 production were studied by using a statistical response surface method. A series of iterative experimental designs was employed to determine optimal fermentation conditions (media and process factors). Optimal ranges characterized by %T-20 (proportion of peptide to the total cell protein) were observed, narrowed down, and further investigated to determine the optimal combination of culture conditions, which was as follows: 9, 6, 10, and 1 mL of glucose, NPK sources, MgSO4, and trace elements, respectively, in a total of 100 mL of medium inducted at an OD of 0.55-0.75 with 0.7 mM isopropyl-beta-D-thiogalactopyranoside in an induction duration of 4 h. Under these conditions, up to 14% of T-20 was obtained. This statistical optimization allowed the production of T-20 to be increased more than twofold (from 6 to 14%) within a shorter induction duration (from 6 to 4 h) at the shake-flask scale.

Statistical medium formulation and process modeling by mixture design of experiment for peptide overexpression in recombinant Escherichia coli.

The medium formulation and robust process modeling for anti-HIV peptide (T-20) production by recombinant Escherichia coli overexpression were studied by employing a crossed experimental design. The crossed design, a mixture design combined with process factor (induction duration), was used to find the optimal medium formulation and process time. The optimal settings for three major components (7.75 mL of NPK sources, 5.5 mL of glucose, and 11.75 mL of MgSO4) characterized by %T-20 (14.45%), the proportion of peptide to the total protein, were observed in a total of 100 mL of medium inducted at an optical density of 0.67 with 0.7 mM isopropyl-beta-D-thiogalactopyranoside) for a 3-h induction duration at shake-flask scale. These conditions were further investigated to find robust process conditions (8.2 mL of NPK sources, 5.6 mL of glucose, and 11.3 mL of MgSO4, and a 3.5-h induction duration time) for T-20 production (13.9%) by applying propagation of error.

Correlation between serum CEA level and metabolic volume as determined by FDG PET in postoperative patients with recurrent colorectal cancer.

To determine the correlation between serum CEA level and the metabolic volume by FDG PET in postoperative patients with recurrent colorectal cancer, FDG PET was performed in 29 consecutive patients with recurrent or metastatic colorectal cancer whose CEA levels were higher than 5 ng/ml. A whole body emission scan was performed 60 minutes after injecting 370-555 MBq of F-18 FDG. "PET volume" and "PET metabolic volume" of tumors were measured on FDG PET images. Based on an isocontour plot of tumor mass at 2.5 SUV (standardized uptake value), the metabolically active tumor "PET volume" was calculated. "PET metabolic volume" was obtained by multiplying the "PET volume" by the mean SUV of the tumor. All recurrent or metastatic lesions were single or multiple lesions of measurable size (axial diameter > 1 cm, minimum "PET volume" 3.5 cm3), and were verified by operation or by other imaging modalities (CT or MRI). There was a linear associations between "PET volume" and serum CEA level. Further regression analysis by least squares showed a highly significant model with an equation of volume = 41.2 + 0.471 x CEA (r2 = 0.629). However, no such association was found between "PET metabolic volume" and serum CEA level according to the residual normality test. In conclusion, "PET volume" measured by FDG PET and serum CEA level in colorectal cancer are significantly correlated. Tumor volume determined by FDG PET can be used as an effective marker of tumor burden in postoperative patients with colorectal carcinoma.

Integration of structural and functional magnetic resonance imaging improves mild cognitive impairment detection.

The identification of mild cognitive impairments (MCI) via either structural magnetic resonance imaging (sMRI) or functional MRI (fMRI) has great potential due to the non-invasiveness of the techniques. Furthermore, these techniques allow longitudinal follow-ups of single subjects via repeated measurements. sMRI- or fMRI-based biomarkers have been adopted separately to diagnose MCI; however, there has not been a systematic effort to integrate sMRI- and fMRI-based features to increase MCI detection accuracy. This study investigated whether the detection of MCI can be improved via the integration of biomarkers identified from both sMRI and fMRI modalities. Regional volume sizes and neuronal activity levels of brains from MCI subjects were compared with those from healthy controls and used to identify biomarkers from sMRI and fMRI data, respectively. In the subsequent classification phase, MCI was automatically detected using a support vector machine algorithm that employed the identified sMRI- and fMRI-based biomarkers as an input feature vector. The results indicate that the fMRI-based biomarkers provided more information for detecting MCI than the sMRI-based biomarkers. Moreover, the integrated feature sets using the sMRI- and fMRI-based biomarkers consistently showed greater detection accuracy than the feature sets based only on the fMRI-based biomarkers. The results demonstrate that integration of sMRI and fMRI modalities can provide supplemental information to improve the diagnosis of MCI relative to either the sMRI or fMRI modalities alone.

Hippocampus-precuneus functional connectivity as an early sign of Alzheimer's disease: a preliminary study using structural and functional magnetic resonance imaging data.

Alzheimer's disease (AD) is characterized by structural atrophies in the hippocampus (HP) and aberrant patterns of functional connectivities (FC) between the hippocampus and the rest of the brain. However, the relationship between cortical atrophy levels and corresponding degrees of aberrant FC patterns has not been systematically examined. In this study, we investigated whether there was an explicit link between structural abnormalities and corresponding functional aberrances associated with AD using structural and functional magnetic resonance imaging (fMRI) data. To this end, brain regions with cortical atrophies that are associated with AD were identified in the HP in the left (L) and right (R) hemispheres using structural MRI data from volume analyses (p<0.03 for L-HP; p<0.04 for R-HP) and voxel-based morphometry analyses (p<4×10(-4) for L-HP; p<2×10(-3) for R-HP). Aberrantly reduced FC levels between the HP (with atrophy) and precuneus were also consistently observed in fMRI data from AD than HC brains that were analyzed by the Pearson's correlation coefficients (p<3×10(-4) for L-HP; and p<8×10(-5) for R-HP). In addition, the substantial negative FC levels from the HC brains between the precuneus and post central gyrus (PoCG) without structural atrophy were also significantly diminished from the AD brains (p<5×10(-5) for L-PoCG; and p<6×10(-5) for R-PoCG). The effect sizes of these aberrant FC levels associated with AD were greater than that of cortical atrophy levels when comparing using normalized Z score and Cohen's d measures, which indicates that an aberrant FC level may precede cortical atrophy.

Multiobjective evolutionary optimization for tumor segmentation of breast ultrasound images.

This paper proposes a robust multiobjective evolutionary algorithm (MOEA) to optimize parameters of tumor segmentation for ultrasound breast images. The proposed algorithm employs efficient schemes for reinforcing proximity to Pareto-optimal and diversity of solutions. They are designed to solve multiobjective problems for segmentation accuracy and speed. First objective is evaluated by difference between the segmented outline and ground truth. Second objective is evaluated by elapsed time during segmentation process. The experimental results show the effectiveness of the proposed algorithm compared with conventional MOEA from the viewpoint of proximity to the Pareto-optimal front (improved by 16.4% and 12.4%). Moreover, segmentation results of proposed algorithm describe faster segmentation speed (1.97 second) and higher accuracy (8% Jaccard).

Mesocorticolimbic hyperactivity of deprived smokers and brain imaging.

We determined the association of neuronal circuitry with the desire to smoke by acquiring and analyzing functional MRI data. The data were acquired in both abstained and subsequently satiated (by 'natural' cigarette smoking) heavy smokers and also in demographically and intellectually matched nonsmokers. During the acquisition, participants were viewing alternating smoking and nonsmoking images that were interleaved by fixation images. From the results, the activities in the mesocorticolimbic pathway including the orbitofrontal cortex, parahippocampus, hippocampus, and midbrain were significantly negatively correlated with carbon monoxide (CO) levels. In contrast, the activities in the motor area and the posterior cingulate cortex plus precuneus were significantly positively correlated with the CO levels. This is the first study to show that mesocorticolimbic and midbrain activities are strongly associated with CO levels, and therefore, possibly with smoking desire levels because of the strong correlation between CO levels and blood nicotine levels.

Real-time fMRI-based neurofeedback reinforces causality of attention networks.

In this study, we investigated the efficacy of a real-time functional magnetic resonance imaging (rtfMRI)-based neurofeedback method for the modulation of the effective connectivity (EC) of causality between attention-related neuronal activities. In participants who received the feedback of attention-related neuronal activity, the EC estimated from Granger causality analysis was reinforced within the task-related network, such as between the bilateral cingulate gyri and frontal cortices, whereas the EC between the task-related network and task-unrelated resting-state network, including the inferior parietal lobule, was diminished. On the other hand, only marginal changes were observed in participants who received "sham" feedback. This "dynamic" characteristic measure of EC based on causality may be useful for evaluating the efficacy of methods designed to modulate brain networks, including rtfMRI-based neurofeedback.

Willingness to pay for hospice care using the contingent valuation method.

PURPOSE: It is necessary to develop a proper payment system for more health care facilities to provide hospice and palliative cares. In deciding the proper level of payment for hospice per diem fee, willingness to pay (WTP) may provide one of the critical information. This study was conducted to determine WTP for hospice care and to analyze those factors affecting WTP. MATERIALS AND METHODS: A contingent valuation method with a double-bounded dichotomous-choice model was used. Interview survey was organized and conducted by a survey company from April 4 to 18, 2008. The mean WTP was calculated through an infinite integration of survival functions. RESULTS: The average willingness to pay was found to be 42,240 Korean won (KRW) (USD 35), with the amount becoming higher as hospice services were deemed more necessary or where average monthly household income was higher. The amount was also higher among male respondents than females. CONCLUSION: To compare this WTP with actual cost (32,500 KRW) (USD 27) for hospice care. To facilitate hospice service, hospice specific payment system should be developed. This study provides information regarding the general public's preference of hospice service and their WTP for hospice care, and it may be useful in the decision-making process.

N-glycosylation status of beta-haptoglobin in sera of patients with prostate cancer vs. benign prostate diseases.

N-glycosylation status of purified beta-haptoglobin separated from sera of patients with prostate cancer was studied in comparison to that of sera from patients with benign prostate diseases, or normal subjects. Two different approaches, as summarized below, one based on binding of lectins and antibodies to beta-haptoglobin, the other on mass spectrometry of released N-linked glycans from beta-haptoglobin, were performed. Some of the results were useful for distinction of prostate cancer vs. benign prostate diseases. i) Binding of Phaseolus vulgaris-L lectin (PHA-L), defining the GlcNAcbeta6Manalpha6Man side chain present in tri- or tetra-antennary N-linked glycans, to beta-haptoglobin was higher for cases of prostate cancer and high-grade prostate intraepithelial neoplasia than for benign diseases. Binding of Aleuria aurantia lectin (AAL) defining Fucalpha3-, alpha4-, or alpha6-GlcNAc, or monoclonal antibody directed to sialyl-Le(x), to beta-haptoglobin was also higher for some of the cancer cases than for benign diseases. Many other lectins and antibodies showed no binding to beta-haptoglobin, or showed no significant difference between cancer vs. benign diseases. ii) Mass spectrometric analysis of N-linked glycans of beta-haptoglobin released by Peptide N-glycosidase-F showed enhanced expression of monosialyl tri-antennary structures in prostate cancer cases. Thus, binding of PHA-L to affinity-purified beta-haptoglobin from sera of patients could lead to development of useful tools for differential diagnosis of prostate cancer vs. benign prostate diseases.

Control of cell motility by interaction of gangliosides, tetraspanins, and epidermal growth factor receptor in A431 versus KB epidermoid tumor cells.

Growth of epidermoid carcinoma cell lines, A431 and KB, has been known to be controlled by the interaction of epidermal growth factor (EGF) and its receptor (EGFR) with tyrosine kinase. Ganglioside GM3 was previously found to interact with EGFR and to inhibit EGFR tyrosine kinase. However, motility of these cells, controlled by EGFR and ganglioside, was not studied. The present study is focused on the control mechanism of the motility of these cells through interaction of ganglioside, tetraspanin (TSP), and EGFR. Key results are as follows: (i) The level of EGFR expressed in A431 cells is approximately 6 times higher than that expressed in KB cells, and motility of A431 cells is also much higher than that of KB cells, yet growth of A431 cells is either not affected or is inhibited by EGF. In contrast, growth of KB cells is enhanced by EGF. (ii) Levels of TSPs (CD9, CD82, and CD81) expressed in A431 cells are much higher than those expressed in KB cells, and TSPs expressed in A431 cells are reduced by treatment of cells with EtDO-P4, which inhibits the synthesis of glycosphingolipids (GSLs) and gangliosides. (iii) These TSPs are co-immunoprecipitated with EGFR in both A431 and KB cells, indicating that TSPs are closely associated with EGFR. (iv) High motility of A431 cells is greatly reduced, while low motility of KB cells is not affected, by treatment of cells with EtDO-P4. These results, taken together, suggest that there is a close correlation between high motility of A431 cells and high expression of EGFR and TSPs, and between ganglioside GM3/GM2 and TSP. A similar correlation was suggested between the low motility of KB cells and low levels of EGFR and TSP. The correlation between high motility and high level of EGFR with the ganglioside-TSP complex in A431 cells is unique. This is in contrast to our previous studies that indicate that motility of many types of tumor cells is inhibited by a high level of CD9 or CD82, together with growth factor receptors and integrins.