The liver is an important site of host-microbe interaction. The liver consists of parenchymal cells (hepatocytes) and nonparenchymal cells (NPC), such as sinusoidal endothelial cells, Kupffer cells, and hepatic stellate cells (1). Kupffer cells that line the liver sinusoid represent the largest fixed population of macrophages in the body and are primarily responsible for endotoxin clearance (2); liver endothelial and parenchymal cells contribute to this process to a lesser extent (3). In addition, NPC (mainly Kupffer cells) secrete potent inflammatory mediators such as reactive oxygen species, eicosanoids, carbon monoxide, tumor necrosis factor-α (TNF-α), and interleukin (IL)-1 and IL-6 (4, 5). IL-1 and TNF-α released by Kupffer cells have been shown to stimulate hepatocyte (HC) inducible nitric oxide (iNO) production (6-8). TNF released by Kupffer cells can induce the hepatic acute phase response or lead to HC apoptosis or necrosis (9). IL-6 is the primary proinflammatory cytokine affecting the acute phase response in the liver (10). IL-6 and TNF-α also have proregenerative effects, with IL-6 also exerting liver protective effects through induction of suppressor of cytokine signaling-3 (SOCS-3) and the antiapoptotic proteins Bcl-2, Bcl-xL, and flice-like inhibitory protein (FLIP) in the liver (11).
Lipopolysaccharide (LPS), a complex glycolipid constituent of the outer membrane of gram-negative bacteria, initiates an intracellular signaling cascade through a transmembrane signaling receptor, Toll-like receptor 4 (TLR4) (12), which ultimately leads to the activation of nuclear factor-κB (NF-κB). Activation of NF-κB results in synthesis and release of cytokines by macrophages. At least three LPS-binding proteins, including LBP (13), CD14 (14), and myeloid differentiation protein 2 (MD2) (15), are required for sensitive and optimal TLR4-mediated LPS recognition. Plasma LBP, an acute phase protein synthesized primarily in the liver, potentiates LPS response by transferring LPS released from the outer membrane of gram-negative bacteria to membrane-bound and soluble forms of CD14 (16). It is commonly believed that Kupffer cells respond to LPS directly and subsequently activate hepatocytes (5), but the mechanism for this effect is not clear. Although LPS activates macrophages through the TLR4-MD2-CD14 complex, it is still uncertain whether this pathway is involved in cytokine production in Kupffer cells. Unlike other LPS-responding macrophages, the constitutive expression of CD14 on Kupffer cells is very low (17), and there is no evidence to show that LPS binds directly to TLR4 on Kupffer cells. As a source of LBP and CD14, we postulated that hepatocytes would augment the responses of the neighboring NPC to LPS.
Using a HC-NPC coculture system, we show that NPC responses to LPS depend on CD14 and TLR4. Furthermore, the presence of hepatocytes increases cytokine production in NPC-HC cocultures. This contact-dependent production of cytokines is also partially dependent on LBP production by hepatocytes.
MATERIALS AND METHODS
LPS (Escherichia coli 0111:B4) was purchased from List Biological Laboratories (Vandell Way, CA). This LPS does not contain a significant amount of contaminating proteins that could stimulate TLR2 nonspecifically (18). Williams Medium E was purchased from Gibco-BRL (Grand Island, NY); fetal calf serum was purchased from HyClone Laboratories (Logan, UT). All tissue culture plates and flasks including Transwell insert were purchased from Corning (Corning, NY). NF-κB consensus oligonucleotides were ordered from Santa Cruz Biotechnology (Santa Cruz, CA).
Experimental protocols were approved by the Institutional Animal Care and Use Committee of the University of Pittsburgh. The experiments were performed in adherence to the National Institutes of Health guidelines on the use of laboratory animals. TLR4-mutant mice (C3H/HeJ) and their control C3H/HeN mice, which were pathogen free and weighed approximately 20 g, were purchased from Jackson Laboratories (Bar Harbor, ME). CD14-deficient mice backcrossed at least six generations on a C57BL/6 genetic background and carrying the CBA-derived NRAMP-1R allele were generated (19) and kindly provided by Douglas T. Golenbock (Boston Medical Center and Boston University School of Medicine, Boston, MA). C57BL/6 mice also bred in our facility were used as wild-type (WT) controls. LBP-deficient mice were housed at the University of Michigan facility together with their WT C57BL/6 controls. The LBP-deficient mice were also originally a generous gift from Douglas T. Golenbock (20). These animals had been backcrossed into the background C57BL/6 strain at least 12 times before use in these experiments. All mice were housed under specific pathogen-free conditions and were exposed each day to 12 h of light and darkness. Rodent chow and water were provided ad libitum.
HC were isolated from mice by an in situ collagenase (type VI; Sigma, St. Louis, MO) perfusion technique, modified as described previously (21). HC were separated from NPC by two cycles of differential centrifugation (50g for 2 min.) and were further purified over a 30% Percoll gradient. The purity of these HC cultures exceeded 98% by light microscopy, and viability was typically over 95% by trypan blue exclusion assay.
Liver NPC were harvested using a modification of the perfusion technique of Seglen (21). The vena cava was cannulated and the liver was perfused for 3 min with 1× Hank's balanced salt solution (Gibco-BRL) supplemented with 0.96 g of sodium bicarbonate/500 mL (Perfusate I) at a flow rate of 10 mL/min. Next, the liver was perfused with a 0.2% protease (Sigma) in Perfusate I (Perfusate II) for 3 min. The liver was dissected out and placed in a petri dish with Perfusate II and diced into 2- to 3-mm pieces. This slurry was then stirred in Perfusate II for 1 h at 37°C. After three washes, the cells were resuspended in culture media and counted. The NPC did not contain HC as detected by light microscopy.
HC and NPC cocultures
The cells were cocultured in Williams Medium E containing 10% calf serum (CS), 15 mM HEPES, 10−6 M insulin, 2 mM L-glutamine, and 100 U/mL penicillin and streptomycin. HC were harvested and plated into 12-well gelatin-coated tissue culture plates (2 × 105/well) and were allowed to attach to plates overnight. NPC were harvested on day 2 and were seeded at a concentration of 1 × 106 cells/mL to give a final HC:NPC ratio of 1:5. HC and NPC from control and mutant or knockout mice were harvested at the same time from age-matched mice and were plated at equivalent concentrations. Cells were initially plated and allowed to adhere using media containing serum. They were then washed extensively followed by replacement with serum-free media before administration of LPS. For each individual experiment, HC or NPC were from a single mouse, with experiments then repeated three to six times. Results from four similar wells of the 12-well plate were combined within each experiment.
Cytokine enzyme-linked immunoabsorbant assay (ELISA) and nitrite assay
TNF-α and IL-6 ELISA kits were used to assay cytokine levels in cell culture supernatants according to the protocols provided by the manufacturer (R&D Systems, Minneapolis, MN). Nitrite concentrations in the culture medium were assessed using the Griess reaction by mixing equal volumes (100 μL) of cell supernatants with the Griess reagent (1% sulfanilamide/0.1% naphthylethylenediamine dihydrochloride/2.5% H3PO4). After incubation at room temperature for 10 min, absorbance was measured spectrophotometrically at 550 nm using a kinetic microplate reader (Vmax; Molecular Devices, Sunnyvale, CA). Nitrite concentration was determined from a calibration curve using sodium nitrite as standard.
Data are presented as mean ± SEM. Experimental results are analyzed for their significance by analysis of variance (ANOVA) and Fisher's PLSD test by statistic software, Statview (Abacus Concepts Inc., Berkeley, CA). Significance was established at the 95% confidence level (P < 0.05).
HC increase NPC cytokine production in response to LPS: the role of TLR4
Liver NPC, especially Kupffer cells, are the major source of cytokine production in the liver during endotoxemia (5). We used HC-NPC cocultures to evaluate the effect of HC-NPC interactions on cytokine production by NPC after LPS stimulation. TNF-α (Fig. 1, A and B) or IL-6 (Fig. 1, C and D) levels in supernatants of HC monocultures are minimal. TNF-α and IL-6 were measured in all cases. Results shown depict TNF-α levels in a time course after LPS, and IL-6 levels showing dose response to LPS. Similar results were obtained for IL-6 production over time after LPS, and production of TNF-α in response to increasing doses of LPS. These data are not shown in the interests of clarity and space saving. As expected, NPC responded to LPS with increased cytokine production. The maximal cytokine level after LPS treatment was observed after 8 h for TNF-α and after 24 h for IL-6. The response of NPC to LPS largely requires functional TLR4 because cytokine production was significantly diminished in NPC isolated from TLR4-mutant (HeJ) mice (Fig. 1, B and D) compared with NPC from TLR4 WT (HeN) mice. When HC from TLR4 WT (HeN) mice were cocultured with autologous NPC in the same dishes, the LPS-induced cytokine levels in culture supernatants were significantly higher than in NPC monocultures. Comparable levels of these cytokines were detected when TLR4 mutant (HeJ) HC were cocultured with NPC from TLR4 WT (HeN) mice, indicating that HC TLR4 was not required for the augmentation of NPC cytokine production. Similar experiments were performed using ratios of HC:NPC of 1:1 and 1:2.5. Results showed a similar pattern of increased cytokine production in cocultures, although because most of the cytokine production is from NPC, the cytokine levels in these samples were lower than when the higher, optimal 1:5 ratio was used (data not shown).
CD14 on HC is not required for augmented cytokine production in HC-NPC cocultures
The response to LPS through TLR4 requires accessory proteins such as CD14 and MD2 to activate NF-κB and proinflammatory cytokine secretion. Therefore, we examined the role of CD14 in cytokine production from HC and NPC monocultures as well as HC-NPC cocultures. HC or NPC were isolated from CD14-deficient (CD14−/−) or WT C57BL/6 mice. IL-6 levels were determined in cell culture supernatants at 4, 8, and 24 h after stimulation with 100 ng/mL LPS (Fig. 2, A and B). CD14 was not required on HC to augment cytokine production in HC-NPC cocultures (Fig. 2A). However, cocultures containing CD14−/− NPC produced lower levels of cytokine compared with cultures of WT NPC, although levels of cytokine production were still augmented in coculture compared with CD14−/− NPC monoculture (Fig. 2B). This indicates that the responses of NPC alone to LPS are CD14 dependent, but that responses of HC-NPC cocultures are only partially CD14 dependent. Furthermore, expression of CD14 by HC is not required for the augmentation effect on NPC. TNF-α levels measured were similarly increased (data not shown).
LBP from HC is important in augmented cytokine production in HC-NPC cocultures
LBP binds LPS and is thought to enhance LPS signaling through the TLR4-receptor complex (13, 16). LBP is mainly expressed as a soluble protein as part of the acute phase response in the liver, but can also be expressed on cells as a membrane-bound protein. Therefore, we investigated the role of HC LBP in interactions with NPC using our coculture system. HC were isolated from LBP-deficient (LBP−/−) or C57BL/6 (WT) mice and were cocultured with WT NPC. TNF-α (Fig. 3A) and IL-6 (Fig. 3B) were again measured in the cell culture supernatants at 4, 8, and 24 h after stimulation with 100 ng/mL LPS. Cocultures with LBP−/− HC had significantly reduced TNF-α and IL-6 levels compared with cocultures containing WT HC, although levels were not reduced to the lower levels of cytokine produced in WT NPC monoculture. Similar experiments were performed using LBP−/− NPC. LBP expression by NPC was not required for increased cytokine production in cocultures (data not shown). Collectively, these data suggest an important role for HC-derived LBP in the augmentation of cytokine levels in HC-NPC coculture.
Close cell proximity is required for HC to augment cytokine release by NPC
To determine if soluble factors produced by HC modulate the response of NPC in this model, we added LPS (to make a final concentration of 100 ng/mL) to fresh serum-free medium or serum-free HC-conditioned medium (HC-CM). HC-CM was serum-free cell culture supernatant from WT HC cell cultures after 36 h. C57BL/6 NPC were then incubated with the control or CM. The HC-CM was diluted serially to eliminate possible inhibitory factors present in the CM. NPC produced equivalent amounts of IL-6 in fresh serum-free medium or HC-CM in response to LPS (Fig. 4A), suggesting that soluble factors released by HC do not contribute to the augmented LPS response by NPC in HC-NPC cocultures.
To determine whether cell contact was required to augment the cytokine levels in HC-NPC coculture, we compared the TNF-α and IL-6 levels in cell culture supernatants from mixed cell cocultures or cocultures separated by Transwell inserts. This system uses a Transwell insert consisting of a semipermeable membrane that allows soluble mediators such as cytokines to diffuse freely, but separates distinct cell populations. HC were allowed to attach to the cell culture plate, with NPC on inserts as appropriate. When HC were cocultured with NPC on the Transwell inserts, the levels of TNF-α (Fig. 4B) and IL-6 (Fig. 4C) in the cell culture supernatants were significantly reduced compared with HC-NPC cocultures, allowing direct cell-to-cell contact. These data suggest that close contact or close cell proximity is required for augmented NPC cytokine production in HC-NPC cocultures. However, this experiment does not rule out the possibility that local production of soluble factors by HC or stimulated by contact with NPC could also augment NPC cytokine production. The presence of the Transwell inserts alone did not significantly alter cytokine responses from HC-NPC cocultures (data not shown).
The liver is an immunocompetent organ that plays a key role in the innate immune response to pathogens. The liver produces inflammatory mediators and acute-phase reactants, and functions to remove pathogens and microbial products from the blood (22, 23). The distinct subtypes of liver cells are arrayed in close proximity to each other. Therefore, HC-NPC interactions are likely to be imperative for coordinated, liver-specific synthetic, metabolic, and detoxification functions (24). The prevailing opinion is that the response of HC to LPS is complex, requiring cell-cell interactions between HC and Kupffer cells, sinusoidal endothelial cell, and stellate cells (24). HC and NPC cocultures have been used to study the roles of HC-NPC interactions (24), including cell survival, proliferation, differentiation, and morphogenesis.
We have reported previously that Kupffer cell-HC cocultures released higher NO reaction products as compared with Kupffer cells alone in response to LPS (25-27). This results from iNOS induction in hepatocytes after cytokine release from Kupffer cells. We have also shown that prostaglandin production is higher in cocultures compared with NPC monocultures (28). Here, we used HC-NPC cocultures to examine cytokine production as well as the role of LPS-signaling molecules TLR4, CD14, and LBP in this process. Our data clearly show that HC-NPC cocultures resulted in augmented cytokine and NO production, and confirm the critical role for TLR4 and CD14 on NPC for optimal response to LPS in the liver. We have also identified the importance of CD14 and LBP from HC in the interactions between HC and NPC.
The augmented production of TNF-α and IL-6 may have implications for endotoxic, or septic, shock. TNF-α released from LPS-treated Kupffer cells contributes to HC apoptosis and necrosis during sepsis, by inhibiting HC mitochondrial function (29), promoting proteolytic HC killing (30), or increasing liver neutrophil accumulation (31). IL-6 from Kupffer cells plays a pleiotropic role in liver regeneration and repair as well as the production of acute-phase proteins during endotoxemia (4), and may exert antiapoptotic effects by maintaining the expression of antiapoptotic proteins such as Bcl-2 and Bcl-xL (11). Our findings confirm a previous study by Hoebe et al. (8) in which HC-NPC cocultures favored TNF-α and IL-6 production by NPC. In addition, we further characterized the critical role for LPS-signaling molecules TLR4, CD14, and LBP in cytokine production by NPC in these cocultures. We have previously shown that mouse HC respond to LPS in a TLR4-dependent manner (32), and Su et al. (33, 34) have shown the critical role for CD14 and TLR4 in NPC (Kupffer cell) responses to LPS. Herein, we reconfirm the observation that NPC from TLR4 mutant and CD14 null mice had diminished TNF-α and IL-6 production (Figs. 1 and 2), and further establish a role for TLR4 and CD14 in NPC in HC-NPC interactions. Our data shows that our observation that cocultures respond to LPS for IL-6 production in the complete absence of CD14 while NPC alone do not is intriguing and raises the possibility that HC could provide LPS-recognition molecules to NPC. LBP on HC is an important factor in augmenting NPC cytokine responses. Because cell contact, or close cell proximity, was required for the augmentation it may be that membrane-bound LBP is required. However, our results do not elucidate the mechanism by which LBP causes the enhanced responses to LPS.
Although we did not use enriched Kupffer cells in our coculture studies, we believe the cytokines produced in our experiments derive mainly from Kupffer cells. This assumption is based on the fact that Kupffer cells represent between 40% and 65% of liver NPC (35) and we have shown very similar responses when purified Kupffer cells have been used instead of NPC (35) (and data not shown). Hence, we conclude that HC are considered not only as targets for Kupffer cell-derived cytokines, but also as active participants of hepatic and systemic cytokine networks. These interactions may modulate hepatic injury or apoptosis during sepsis.
1. Kmiec Z: Cooperation of liver cells in health and disease. Adv Anat Embryol Cell Biol
2. Ruiter DJ, van der Meulen J, Brouwer A, Hummel MJ, Mauw BJ, van der Ploeg JC, Wisse E: Uptake by liver cells of endotoxin following its intravenous injection. Lab Invest
3. Fox ES, Thomas P, Broitman SA: Uptake and modification of 125
I-lipopolysaccharide by isolated rat Kupffer cells. Hepatology
4. Streetz KL, Wustefeld T, Klein C, Manns MP, Trautwein C: Mediators of inflammation and acute phase response in the liver. Cell Mol Biol
5. Su GL: Lipopolysaccharides in liver injury: molecular mechanisms of Kupffer cell activation. Am J Physiol Gastrointest Liver Physiol
6. de Vera ME, Wong JM, Zhou JY, Tzeng E, Wong HR, Billiar TR, Geller DA: Cytokine-induced nitric oxide synthase gene transcription is blocked by the heat shock response in human liver cells. Surgery
7. Geller DA, Freeswick PD, Nguyen D, Nussler AK, Di Silvio M, Shapiro RA, Wang SC, Simmons RL, Billiar TR: Differential induction of nitric oxide synthase in hepatocytes during endotoxemia and the acute-phase response. Arch Surg
8. Hoebe KH, Witkamp RF, Fink-Gremmels J, Van Miert AS, Monshouwer M: Direct cell-to-cell contact between Kupffer cells and hepatocytes augments endotoxin-induced hepatic injury. Am J Physiol Gastrointest Liver Physiol
9. Zhao Y, Li S, Childs EE, Kuharsky DK, Yin XM: Activation of pro-death Bcl-2 family proteins and mitochondria apoptosis pathway in tumor necrosis factor-α-induced liver injury. J Biol Chem
10. Ramadori G, Christ B: Cytokines
and the hepatic acute-phase response. Semin Liver Dis
11. Kovalovich K, Li W, DeAngelis R, Greenbaum LE, Ciliberto G, Taub R: Interleukin-6 protects against Fas-mediated death by establishing a critical level of anti-apoptotic hepatic proteins FLIP, Bcl-2, and Bcl-xL. J Biol Chem
12. Poltorak A, He X, Smirnova I, Liu MY, Van Huffel C, Du X, Birdwell D, Alejos E, Silva M, Galanos C, Freudenberg M, Ricciardi-Castagnoli P, Layton B, Beutler B: Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4
13. Jack RS, Fan X, Bernheiden M, Rune G, Ehlers M, Weber A, Kirsch G, Mentel R, Furll B, Freudenberg M, Schmitz G, Stelter F, Schutt C: Lipopolysaccharide-binding protein is required to combat a murine gram-negative bacterial infection. Nature
14. Haziot A, Ferrero E, Kontgen F, Hijiya N, Yamamoto S, Silver J, Stewart CL, Goyert SM: Resistance to endotoxin shock and reduced dissemination of gram-negative bacteria in CD14
-deficient mice. Immunity
15. Nagai Y, Akashi S, Nagafuku M, Ogata M, Iwakura Y, Akira S, Kitamura T, Kosugi A, Kimoto M, Miyake K: Essential role of MD-2 in LPS responsiveness and TLR4
distribution. Nat Immunol
16. Wright SD, Ramos RA, Tobias PS, Ulevitch RJ, Mathison JC: CD14
, a receptor for complexes of lipopolysaccharide (LPS) and LPS binding protein. Science
17. Takai N, Kataoka M, Higuchi Y, Matsuura K, Yamamoto S: Primary structure of rat CD14
and characteristics of rat CD14
, cytokine, and NO synthase mRNA expression in mononuclear phagocyte system cells in response to LPS. J Leukocyte Biol
18. Hirschfeld M, Ma Y, Weis JH, Vogel SN, Weis JJ: Cutting edge: repurification of lipopolysaccharide eliminates signaling through both human and murine toll-like receptor 2. J Immunol
19. Moore KJ, Andersson LP, Ingalls RR, Monks BG, Li R, Arnaout MA, Golenbock DT, Freeman MW: Divergent response to LPS and bacteria in CD14
-deficient murine macrophages. J Immunol
20. Wurfel MM, Monks BG, Ingalls RR, Dedrick RL, Delude R, Zhou D, Lamping N, Schumann RR, Thieringer R, Fenton MJ, Wright SD, Golenbock D: Targeted deletion of lipopolysaccharide (LPS)-binding protein gene leads to profound suppression of LPS responses ex vivo, whereas in vivo responses remain intact. J Exp Med
21. Seglen PO: Preparation of isolated rat liver cells. Methods Cell Biol
22. Tanikawa K, Mimura Y, Sakisaka S, Noguchi K: Role of hepatocytes in the clearance of lipopolysaccharide and its clinical significance. Prog Clin Biol Res
23. Vodovotz Y, Liu S, McCloskey C, Shapiro R, Green A, Billiar TR: The hepatocyte as a microbial product-responsive cell. J Endotoxin Res
24. Bhatia SN, Balis UJ, Yarmush ML, Toner M: Effect of cell-cell interactions in preservation of cellular phenotype: cocultivation of hepatocytes and nonparenchymal cells. FASEB J
25. Billiar TR, Curran RD, Ferrari FK, Williams DL, Simmons RL: Kupffer cell:hepatocyte co-cultures release nitric oxide in response to bacterial endotoxin. J Surg Res
26. Billiar TR, Curran RD: Kupffer cell and hepatocyte interactions: a brief overview. J Parenter Enteral Nutr
27. Curran RD, Billiar TR, Steuhr DJ, Hofmann K, Simmons RL: Hepatocytes produce nitrogen oxides from L-arginine in response to inflammatory products of Kupffer cells. J Exp Med
28. Billiar TR, Lysz TW, Curran RD, Bentz BG, Machiedo GW, Simmons RL: Hepatocyte modulation of Kupffer cell prostaglandin E2 production in vitro. J Leukocyte Biol
29. Kurose I, Miura S, Higuchi H, Watanabe N, Kamegaya Y, Takaishi M, Tomita K, Fukumura D, Kato S, Ishii H: Increased nitric oxide synthase activity as a cause of mitochondrial dysfunction in rat hepatocytes: roles for tumor necrosis factor α. Hepatology
30. Hamada E, Nishida T, Uchiyama Y, Nakamura J, Isahara K, Kazuo H, Huang TP, Momoi T, Ito T, Matsuda H: Activation of Kupffer cells and caspase-3 involved in rat hepatocyte apoptosis induced by endotoxin. J Hepatol
31. Sakamoto S, Okanoue T, Itoh Y, Nakagawa Y, Nakamura H, Morita A, Daimon Y, Sakamoto K, Yoshida N, Yoshikawa T, Kashima K: Involvement of Kupffer cells in the interaction between neutrophils and sinusoidal endothelial cells in rats. Shock
32. Liu S, Gallo DJ, Green AM, Williams DL, Gong X, Shapiro RA, Gambotto AA, Humphris EL, Vodovotz Y, Billiar TR: Role of toll-like receptors in changes in gene expression and NF-κB activation in mouse hepatocytes stimulated with lipopolysaccharide. Infect Immun
33. Su GL, Klein RD, Aminlari A, Zhang HY, Steinstraesser L, Alarcon WH, Remick DG, Wang SC: Kupffer cell activation by lipopolysaccharide in rats: role for lipopolysaccharide binding protein and toll-like receptor 4. Hepatology
34. Su GL, Goyert SM, Fan MH, Aminlari A, Gong KQ, Klein RD, Myc A, Alarcon WH, Steinstraesser L, Remick DG, Wang SC: Activation of human and mouse Kupffer cells by lipopolysaccharide is mediated by CD14
. Am J Physiol Gastrointest Liver Physiol
35. Steinhorn DM, Cerra FB: Comparative effects of lipopolysaccharide on newborn versus adult rat hepatocyte and nonparenchymal cell co-cultures. Crit Care Med