Molecular interplay between HURP and Kif18A in mitotic spindle regulation.

During mitosis, microtubule dynamics are regulated to ensure proper alignment and segregation of chromosomes. The dynamics of kinetochore-attached microtubules are regulated by hepatoma-upregulated protein (HURP) and the mitotic kinesin-8 Kif18A, but the underlying mechanism remains elusive. Using single-molecule imaging in vitro, we demonstrate that Kif18A motility is regulated by HURP. While sparse decoration of HURP activates the motor, higher concentrations hinder processive motility. To shed light on this behavior, we determined the binding mode of HURP to microtubules using Cryo-EM. The structure reveals that one HURP motif spans laterally across ß-tubulin, while a second motif binds between adjacent protofilaments. HURP partially overlaps with the microtubule-binding site of the Kif18A motor domain, indicating that excess HURP inhibits Kif18A motility by steric exclusion. We also observed that HURP and Kif18A function together to suppress dynamics of the microtubule plus-end, providing a mechanistic basis for how they collectively serve in spindle length control.

Molecular interplay between HURP and Kif18A in mitotic spindle regulation.

During mitosis, microtubule dynamics are regulated to ensure proper alignment and segregation of chromosomes. The dynamics of kinetochore-attached microtubules are regulated by hepatoma-upregulated protein (HURP) and the mitotic kinesin-8 Kif18A, but the underlying mechanism remains elusive. Using single-molecule imaging in vitro , we demonstrate that Kif18A motility is regulated by HURP. While sparse decoration of HURP activates the motor, higher concentrations hinder processive motility. To shed light on this behavior, we determined the binding mode of HURP to microtubules using Cryo-EM. The structure reveals that one HURP motif spans laterally across ß-tubulin, while a second motif binds between adjacent protofilaments. HURP partially overlaps with the microtubule-binding site of the Kif18A motor domain, indicating that excess HURP inhibits Kif18A motility by steric exclusion. We also observed that HURP and Kif18A function together to suppress dynamics of the microtubule plus-end, providing a mechanistic basis for how they collectively serve in spindle length control.

Lis1 slows force-induced detachment of cytoplasmic dynein from microtubules.

Lis1 is a key cofactor for the assembly of active cytoplasmic dynein complexes that transport cargo along microtubules. Lis1 binds to the AAA+ ring and stalk of dynein and slows dynein motility, but the underlying mechanism has remained unclear. Using single-molecule imaging and optical trapping assays, we investigated how Lis1 binding affects the motility and force generation of yeast dynein in vitro. We showed that Lis1 slows motility by binding to the AAA+ ring of dynein, not by serving as a roadblock or tethering dynein to microtubules. Lis1 binding also does not affect force generation, but it induces prolonged stalls and reduces the asymmetry in the force-induced detachment of dynein from microtubules. The mutagenesis of the Lis1-binding sites on the dynein stalk partially recovers this asymmetry but does not restore dynein velocity. These results suggest that Lis1-stalk interaction slows the detachment of dynein from microtubules by interfering with the stalk sliding mechanism.

Dyneins.

Dyneins are a family of motor proteins that carry out motility and force generation functions towards the minus end of microtubule filaments. Cytoplasmic dynein (dynein-1) is responsible for transporting intracellular cargos in the retrograde direction in the cytoplasm, anchoring several organelles to the microtubule network, driving nuclear migration in developing neurons, and orienting the mitotic spindle in dividing cells. All other dyneins are localized to cilia. Similar to dynein-1, dynein-2 walks along microtubules and drives intraflagellar transport in the retrograde direction. Other ciliary dyneins are positioned between adjacent microtubule doublets of the axoneme and power ciliary beating by sliding microtubules relative to each other. In this primer, we first highlight the structure, mechanism, and regulation of dynein-1, which is the best-characterized member of the dynein motor family, and then describe the unique features and cellular roles of other dyneins. We also discuss accessory proteins that regulate the activation and motility of dynein motors in different cellular contexts.

Nde1 promotes Lis1-mediated activation of dynein.

Cytoplasmic dynein drives the motility and force generation functions towards the microtubule minus end. The assembly of dynein with dynactin and a cargo adaptor in an active transport complex is facilitated by Lis1 and Nde1/Ndel1. Recent studies proposed that Lis1 relieves dynein from its autoinhibited conformation, but the physiological function of Nde1/Ndel1 remains elusive. Here, we investigate how human Nde1 and Lis1 regulate the assembly and subsequent motility of mammalian dynein using in vitro reconstitution and single molecule imaging. We find that Nde1 recruits Lis1 to autoinhibited dynein and promotes Lis1-mediated assembly of dynein-dynactin adaptor complexes. Nde1 can compete with the α2 subunit of platelet activator factor acetylhydrolase 1B (PAF-AH1B) for the binding of Lis1, which suggests that Nde1 may disrupt PAF-AH1B recruitment of Lis1 as a noncatalytic subunit, thus promoting Lis1 binding to dynein. Before the initiation of motility, the association of dynactin with dynein triggers the dissociation of Nde1 from dynein by competing against Nde1 binding to the dynein intermediate chain. Our results provide a mechanistic explanation for how Nde1 and Lis1 synergistically activate the dynein transport machinery.

Development of a Ratiometric Tension Sensor Exclusively Responding to Integrin Tension Magnitude in Live Cells.

Integrin tensions are critical for cell mechanotransduction. By converting force to fluorescence, molecular tension sensors image integrin tensions in live cells with a high resolution. However, the fluorescence signal intensity results collectively from integrin tension magnitude, tension dwell time, integrin density, sensor accessibility, and so forth, making it highly challenging to specifically monitor the molecular force level of integrin tensions. Here, a ratiometric tension sensor (RTS) was developed to exclusively monitor the integrin tension magnitude. The RTS consists of two tension-sensing units that are coupled in series and always subject to the same integrin tension. These two units are activated by tension to fluoresce in separate spectra and with different activation rates. The ratio of their activation probabilities, reported by fluorescence ratiometric measurement, is solely determined by the local integrin tension magnitude. RTS responded sensitively to the variation of integrin tension magnitude in platelets and focal adhesions due to different cell plating times, actomyosin inhibition, or vinculin knockout. At last, RTS confirmed that integrin tension magnitude in platelets and focal adhesions decreases monotonically with the substrate rigidity, verifying the rigidity dependence of integrin tensions in live cells and suggesting that integrin tension magnitude could be a key biomechanical factor in cell rigidity sensing.

Nde1 Promotes Lis1-Mediated Activation of Dynein.

Cytoplasmic dynein is the primary motor that drives the motility and force generation functions towards the microtubule minus end. The activation of dynein motility requires its assembly with dynactin and a cargo adaptor. This process is facilitated by two dynein-associated factors, Lis1 and Nde1/Ndel1. Recent studies proposed that Lis1 rescues dynein from its autoinhibited conformation, but the physiological function of Nde1/Ndel1 remains elusive. Here, we investigated how human Nde1 and Lis1 regulate the assembly and subsequent motility of the mammalian dynein/dynactin complex using in vitro reconstitution and single molecule imaging. We found that Nde1 promotes the assembly of active dynein complexes in two distinct ways. Nde1 competes with the α2 subunit of platelet activator factor acetylhydrolase (PAF-AH) 1B, which recruits Lis1 as a noncatalytic subunit and prevents its binding to dynein. Second, Nde1 recruits Lis1 to autoinhibited dynein and promotes Lis1-mediated assembly of dynein-dynactin-adaptor complexes. However, excess Nde1 inhibits dynein, presumably by competing against dynactin to bind the dynein intermediate chain. The association of dynactin with dynein triggers Nde1 dissociation before the initiation of dynein motility. Our results provide a mechanistic explanation for how Nde1 and Lis1 synergistically activate the dynein transport machinery.

Ubiquitous membrane-bound DNase activity in podosomes and invadopodia.

Podosomes and invadopodia, collectively termed invadosomes, are adhesive and degradative membrane structures formed in many types of cells and are well known for recruiting various proteases. However, another major class of degradative enzymes, deoxyribonuclease (DNase), remains unconfirmed and not studied in invadosomes. Here, using surface-immobilized nuclease sensor (SNS), we demonstrated that invadosomes recruit DNase to their core regions, which degrade extracellular double-stranded DNA. We further identified the DNase as GPI-anchored membrane-bound DNase X. DNase recruitment is ubiquitous and consistent in invadosomes of all tested cell types. DNase activity exhibits within a minute after actin nucleation, functioning concomitantly with protease in podosomes but preceding it in invadopodia. We further showed that macrophages form DNase-active podosome rosettes surrounding bacteria or micropatterned antigen islets, and the podosomes directly degrade bacterial DNA on a surface, exhibiting an apparent immunological function. Overall, this work reports DNase in invadosomes for the first time, suggesting a richer arsenal of degradative enzymes in invadosomes than known before.

Cell Migration Driven by Self-Generated Integrin Ligand Gradient on Ligand-Labile Surfaces.

Integrin-ligand interaction mediates the adhesion and migration of many metazoan cells. Here, we report a unique mode of cell migration elicited by the lability of integrin ligands. We found that stationary cells spontaneously turn migratory on substrates where integrin ligands are subject to depletion by cellular force. Using TGT, a rupturable molecular linker, we quantitatively tuned the rate of ligand rupture by cellular force and tested platelets (anucleate cells), CHO-K1 cells (nucleated cells), and other cell types on TGT surfaces. These originally stationary cells readily turn motile on the uniform TGT surface, and their motility is correlated with the ligand depletion rate caused by cells. We named this new migration mode ligand-depleting (LD) migration. Through both experiments and simulations, we revealed the biophysical mechanism of LD migration. We found that the cells create and maintain a gradient of ligand surface density underneath the cell body by constantly rupturing local ligands, and the gradient in turn drives and guides cell migration. This is reminiscent of the phenomenon that some liquid droplets or solid beads can spontaneously move on homogeneous surfaces by chemically forming and maintaining a local gradient of surface energy. Here, we showed that cells, as living systems, can harness a similar mechanism to migrate. LD migration is beneficial for cells to maintain adhesion on ligand-labile surfaces, and might also play a role in the migration of cancer cells, immune cells, and platelets that deplete adhesive ligands of the matrix.

Cellular Force Nanoscopy with 50 nm Resolution Based on Integrin Molecular Tension Imaging and Localization.

Integrin-transmitted cellular forces have rich spatial dynamics and are vital to many cellular functions. To advance the sensitivity and spatial resolution of cellular force imaging, we developed a force-activatable emitter reporting single-molecular tension events and the associated cellular force nanoscopy (CFN). Immobilized on a surface, the emitters are initially dark (>99.8% quenched), providing a low fluorescence background despite the high coating density (>2000/µm2) required for sampling cellular force properly. The emitters fluoresce brightly once switched on by integrin tensions and can be switched off by photobleaching, enabling continuous real-time imaging of integrin molecular tensions in live cells. With multiple cycles of molecular tension imaging and localization, CFN reproduces cellular force images with 50 nm resolution. Applied to both migratory cells and stationary cells, CFN revealed ultranarrow distribution of integrin tensions at the cell leading edge, and showed that force distribution in focal adhesions (FAs) is off-centered and FA size-dependent.

Peptide nucleic acid based tension sensor for cellular force imaging with strong DNase resistance.

DNA is a versatile biomaterial with well-defined mechanical and biochemical properties. It has been broadly adopted to synthesize tension sensors that calibrate and visualize cellular forces at the cell-matrix interface. Here we showed that DNA-based tension sensors are vulnerable to deoxyribonucleases (DNases) which cells may express on cell membrane or secret to the culture environment. These DNases can damage the sensors, lower signal-to-noise ratio or even produce false signal in cellular force imaging. To address this issue, we tested peptide nucleic acid (PNA), chemically modified RNA and their hybrids with DNA as alternative biomaterials for constructing tension sensors. Four duplexes: double-stranded DNA (dsDNA), PNA/DNA, dsRNA (modified RNA) and PNA/RNA, were tested and evaluated in terms of DNase resistance, cellular force imaging ability and material robustness. The results showed that all PNA/DNA, dsRNA and PNA/RNA exhibited strong resistance to both soluble DNase I and membrane-bound DNase on cells. However, PNA/RNA-based tension sensor had low signal-to-noise ratio in cellular force imaging, and dsRNA-based tension sensor exhibited strong non-specific signal unrelated to cellular forces. Only PNA/DNA-based tension sensor reported cellular forces with highest signal-to-noise ratio and specificity. Collectively, we confirmed that PNA/DNA hybrid is an accessible material for the synthesis of DNase-resistant tension sensor that retains the force-reporting capability and remains stable in DNase-expressing cells. This new class of tension sensors will broaden the application of tension sensors in the study of cell mechanobiology.

Imaging Integrin Tension and Cellular Force at Submicron Resolution with an Integrative Tension Sensor.

Molecular tension transmitted by integrin-ligand bonds is the fundamental mechanical signal in the integrin pathway that plays significant roles in many cell functions and behaviors. To calibrate and image integrin tension with high force sensitivity and spatial resolution, we developed an integrative tension sensor (ITS), a DNA-based fluorescent tension sensor. The ITS is activated to fluoresce if sustaining a molecular tension, thus converting force to fluorescent signal at the molecular level. The tension threshold for ITS activation is tunable in the range of 10-60 pN that well covers the dynamic range of integrin tension in cells. On a substrate grafted with an ITS, the integrin tension of adherent cells is visualized by fluorescence and imaged at submicron resolution. The ITS is also compatible with cell structural imaging in both live cells and fixed cells. The ITS has been successfully applied to the study of platelet contraction and cell migration. This paper details the procedure for the synthesis and application of the ITS in the study of integrin-transmitted cellular force.

Optical sensor revealed abnormal nuclease spatial activity on cancer cell membrane.

Nucleases are important enzymes that cleave nucleic acids and play critical roles in DNA repair, immune defense and potentially in cancer invasion. However, their spatial dynamics at subcellular level is much less studied. Here, we developed a surface-tethered nuclease sensor (SNS) which directly converts membrane-bound nuclease (MN) activity to fluorescent signal, therefore, mapping MN activity on cell adhesion sites with high resolution and sensitivity. With SNS, we studied MN activity on the ventral membrane of cancer cells, where MN activity initially occurs in punctate regions and advances in a coral-shaped pattern. In six tested cell-lines, the MN activity levels in cancer cells are significantly higher than those in non-cancer cells. We then tested SNS as a sensitive approach to detect cancer cells at single cell level. Single breast cancer cells were successfully detected from thousands of adherent non-cancer cells and from millions of non-adherent blood cells.

Keratocytes Generate High Integrin Tension at the Trailing Edge to Mediate Rear De-adhesion during Rapid Cell Migration.

Rapid cell migration requires efficient rear de-adhesion. It remains undetermined whether cells mechanically detach or biochemically disassemble integrin-mediated rear adhesion sites in highly motile cells such as keratocytes. Using molecular tension sensor, we calibrated and mapped integrin tension in migrating keratocytes. Our experiments revealed that high-level integrin tension abbreviated as HIT, in the range of 50-100 pN (piconewton) and capable of rupturing integrin-ligand bonds, is exclusively and narrowly generated at cell rear margin during cell migration. Co-imaging of HIT and focal adhesions (FAs) shows that HIT is produced to mechanically peel off FAs that lag behind, and HIT intensity is correlated with the local cell retraction rate. High-level molecular tension was also consistently generated at the cell margin during artificially induced cell front retraction and during keratocyte migration mediated by biotin-streptavidin bonds. Collectively, these experiments provide direct evidence showing that migrating keratocytes concentrate force at the cell rear margin to mediate rear de-adhesion.

Force-activatable coating enables high-resolution cellular force imaging directly on regular cell culture surfaces.

Integrin-transmitted cellular forces are crucial mechanical signals regulating a vast range of cell functions. Although various methods have been developed to visualize and quantify cellular forces at the cell-matrix interface, a method with high performance and low technical barrier is still in demand. Here we developed a force-activatable coating (FAC), which can be simply coated on regular cell culture apparatus' surfaces by physical adsorption, and turn these surfaces to force reporting platforms that enable cellular force mapping directly by fluorescence imaging. The FAC molecule consists of an adhesive domain for surface coating and a force-reporting domain which can be activated to fluoresce by integrin molecular tension. The tension threshold required for FAC activation is tunable in 10-60 piconewton (pN), allowing the selective imaging of cellular force contributed by integrin tension at different force levels. We tested the performance of two FACs with tension thresholds of 12 and 54 pN (nominal values), respectively, on both glass and polystyrene surfaces. Cellular forces were successfully mapped by fluorescence imaging on all the surfaces. FAC-coated surfaces also enable co-imaging of cellular forces and cell structures in both live cells and immunostained cells, therefore opening a new avenue for the study of the interplay of force and structure. We demonstrated the co-imaging of integrin tension and talin clustering in live cells, and concluded that talin clustering always occurs before the generation of integrin tension above 54 pN, reinforcing the notion that talin is an important adaptor protein for integrin tension transmission. Overall, FAC provides a highly convenient approach that is accessible to general biological laboratories for the study of cellular forces with high sensitivity and resolution, thus holding the potential to greatly boost the research of cell mechanobiology.

Force-activatable biosensor enables single platelet force mapping directly by fluorescence imaging.

Integrin-transmitted cellular forces are critical for platelet adhesion, activation, aggregation and contraction during hemostasis and thrombosis. Measuring and mapping single platelet forces are desired in both research and clinical applications. Conventional force-to-strain based cell traction force microscopies have low resolution which is not ideal for cellular force mapping in small platelets. To enable platelet force mapping with submicron resolution, we developed a force-activatable biosensor named integrative tension sensor (ITS) which directly converts molecular tensions to fluorescent signals, therefore enabling cellular force mapping directly by fluorescence imaging. With ITS, we mapped cellular forces in single platelets at 0.4µm resolution. We found that platelet force distribution has strong polarization which is sensitive to treatment with the anti-platelet drug tirofiban, suggesting that the ITS force map can report anti-platelet drug efficacy. The ITS also calibrated integrin molecular tensions in platelets and revealed two distinct tension levels: 12-54 piconewton (nominal values) tensions generated during platelet adhesion and tensions above 54 piconewton generated during platelet contraction. Overall, the ITS is a powerful biosensor for the study of platelet mechanobiology, and holds great potential in antithrombotic drug development and assessing platelet activity in health and disease.

Glycine protects against high sucrose and high fat-induced non-alcoholic steatohepatitis in rats.

We set out to explore the hypothesis that glycine attenuates non-alcoholic steatohepatitis (NASH) in rats and the possible mechanism by which is it does. Male Sprague-Dawley (SD) rats were fed a diet containing high fat and high sucrose (HSHF) for 24 weeks to induce NASH. Blood and liver tissues were sampled at selected time points throughout the study. Compared with control animals, the content of alanine transaminase (ALT), triglycerides (TGs), and free fatty acids (FFAs) in plasma and the TG and FFA content in the liver was increased from week 4 to 24. The level of TNFα and MCP-1 in plasma, the content of TNFα in the liver, the insulin resistance index, inflammatory cell infiltration, hepatocyte apoptosis, reactive oxygen species (ROS) generation, and endoplasmic stress-associated protein expression were unaltered at 4 weeks. However, these levels were significantly elevated in HSHF fed rats at 12 weeks. At the same time, the level of endotoxin progressively increased from 0.08 ± 0.02 endotoxin EU/ml at week 4 to 0.7 ± 0.19 EU/ml at week 24. Moreover, these rats had elevated blood endotoxin levels, which were positively associated with their NASH indexes. Liver histology progressively worsened over the course of the study. However, we found that with concomitant treatment with glycine, the level of endotoxin decreased, while NASH indexes significantly decreased and liver status markedly improved,. These data support the hypothesis that glycine protects against NASH in rats by decreasing the levels of intestinal endotoxin, alleviating endoplasmic reticulum and oxidative stress.

A model of metabolic syndrome and related diseases with intestinal endotoxemia in rats fed a high fat and high sucrose diet.

AIM: We sought develop and characterize a diet-induced model of metabolic syndrome and its related diseases. METHODS: The experimental animals (Spague-Dawley rats) were randomly divided into two groups, and each group was fed a different feed for 48 weeks as follows: 1) standard control diet (SC), and 2) a high sucrose and high fat diet (HSHF). The blood, small intestine, liver, pancreas, and adipose tissues were sampled for analysis and characterization. RESULTS: Typical metabolic syndrome (MS), non-alcoholic fatty liver disease (NAFLD), and type II diabetes (T2DM) were common in the HSHF group after a 48 week feeding period. The rats fed HSHF exhibited signs of obesity, dyslipidemia, hyperglycaemia, glucose intolerance, and insulin resistance (IR). At the same time, these animals had significantly increased levels of circulating LPS, TNFα, and IL-6 and increased ALP in their intestinal tissue homogenates. These animals also showed a significant reduction in the expression of occluding protein. The HSHF rats showed fatty degeneration, inflammation, fibrosis, cirrhosis, and lipid accumulation when their liver pathologies were examined. The HSHF rats also displayed increased islet diameters from 12 to 24 weeks, while reduced islet diameters occurred from 36 to 48 weeks with inflammatory cell infiltration and islet fat deposition. The morphometry of adipocytes in HSHF rats showed hypertrophy and inflammatory cell infiltration. HSHF CD68 analysis showed macrophage infiltration and significant increases in fat and pancreas size. HSHF Tunel analysis showed significant increases in liver and pancreas cell apoptosis. CONCLUSIONS: This work demonstrated the following: 1) a characteristic rat model of metabolic syndrome (MS) can be induced by a high sucrose and high fat diet, 2) this model can be used to research metabolic syndrome and its related diseases, such as NAFLD and T2DM, and 3) intestinal endotoxemia (IETM) may play an important role in the pathogenesis of MS and related diseases, such as NAFLD and T2DM.

[Changes in mast cells and hepatic expression of c-kit and stem cell factor in the rat model of chronic hepatitis].

OBJECTIVE: To study the potential role of mast cells and the related molecular mechanism in chronic hepatitis (CH) using a rat model system. METHODS: Thirty Wistar rats (15 males, 15 females; weight range: 230-290 g) were randomly divided into the normal contrast (NC) group and experimental CH group. The CH group received subcutaneous injection of CCl4 and a diet high in cholesterol and alcohol content and low in protein and choline content. Throughout the 4-week modeling period, aseptic blood samples were taken to test plasma tryptase (TS) and hyaluronic acid (HA) levels. The rats were euthanized to assess the changes in liver mast cells by histology and morphology analyses and the changes in liver expression of c-kit and stem cell factor (SCF) proteins by immunohistochemistry and mRNAs by RT-PCR. RESULTS: Compared to the NC group, the CH group had higher plasma and liver concentration of HA (78.09 +/- 38.55 vs. 145.14 +/- 52.54 ng/ml, 51.58 +/- 20.45 vs. 106.59 +/- 43.15 ng/100 mg; t = 2.457 and 2.825 respectively, both P less than 0.05) and TS (0.416 +/- 0.143 vs 0.753 +/- 0.210 mg/ml; t = 4.165, P less than 0.05). The CH group also showed fatty degeneration and fibrosis with many degranulating and degranulated mast cells filled with purple granula located around the liver blood vessels and in fiber-intervals. The CH livers also showed a significantly higher number of mast cells (2.167 +/- 0.924 vs. NC: 10.92 +/- 1.575; t = 7.633, P less than 0.05) and stronger intensity of c-kit staining (2.783 +/- 0.577 vs. 12.86 +/- 3.126; t = 9.511, P less than 0.05) and SCF staining (3.383 +/- 1.583 vs. 15.58 +/- 6.431; t = 9.625, P less than 0.05). The expressions of c-kit and SCF were positively correlated with HA level (r = 0.478 and 0.556 respectively, both P less than 0.05). The c-kit and SCF mRNA expression levels were also significantly higher in the CH liver tissues. CONCLUSION: Mast cell degranulation and histamine release is significantly increased under conditions of chronic hepatitis, and the related mechanism may involve up-regulation of the membrane receptor c-kit and its ligand SCF.

Effects of glycine on phagocytosis and secretion by Kupffer cells in vitro.

AIM: To investigate the effects and mechanisms of action of glycine on phagocytosis and tumor necrosis factor (TNF)-α secretion by Kupffer cells in vitro. METHODS: Kupffer cells were isolated from normal rats by collagenase digestion and Percoll density gradient differential centrifugation. After culture for 24 h, Kupffer cells were incubated in fresh Dulbecco's Modification of Eagle's Medium containing glycine (G1: 1 mmol/L, G2: 10 mmol/L, G3: 100 mmol/L and G4: 300 mmol/L) for 3 h, then used to measure phagocytosis by a bead test, TNF-α secretion after lipopolysaccharide stimulation by radioactive immunoassay, and microfilament and microtubule expression by staining with phalloidin-fluorescein isothiocyanate (FITC) or a monoclonal anti-α tubulin-FITC antibody, respectively, and evaluated under a ultraviolet fluorescence microscope. RESULTS: Glycine decreased the phagocytosis of Kupffer cells at both 30 min and 60 min (P < 0.01, P < 0.05). The numbers of beads phagocytosed by Kupffer cells in 30 min were 16.9 ± 4.0 (control), 9.6 ± 4.1 (G1), 12.1 ± 5.7 (G2), 8.1 ± 3.2 (G3) and 7.5 ± 2.0 (G4), and were 22.5 ± 7.9 (control), 20.1 ± 5.8 (G1), 19.3 ± 4.8 (G2), 13.5 ± 4.7 (G3) and 9.2 ± 3.1 (G4) after 60 min. TNF-α secretion by Kupffer cells in G1 (0.19 ± 0.03), G2 (0.16 ± 0.04), G3 (0.14 ± 0.03) and G4 (0.13 ± 0.05) was significantly less than that in controls (0.26 ± 0.03, P < 0.01), and the decrease in secretion was dose-dependent (P < 0.05). Microfilaments of Kupffer cells in G2, G3 and G4 groups were arranged in a disorderly manner. The fluorescence densities of microtubules in G1 (53.4 ± 10.5), G2 (54.1 ± 14.6), G3 (64.9 ± 12.1) and G4 (52.1 ± 14.2) were all lower than those in the controls (102.2 ± 23.7, P < 0.01), but the decrease in microtubule fluorescence density was not dose-dependant. CONCLUSION: Glycine can decrease the phagocytosis and secretion by Kupffer cells in vitro, which may be related to the changes in the expression of microfilaments and microtubules induced by Kupffer cells.