Basic Science: Concise Communication
Exposure to HIV-1-encoded Toll-like receptor 8 ligands enhances monocyte response to microbial encoded Toll-like receptor 2/4 ligands
Mureith, Marianne Wa,b; Chang, J Judyb; Lifson, Jeffrey Dc; Ndung'u, Thumbia; Altfeld, Marcusa,b
aHIV Pathogenesis Programme, Doris Duke Medical Research Institute and KwaZulu-Natal Research Institute for TB and HIV, Nelson R. Mandela Medical School, University of KwaZulu-Natal, Durban, South Africa
bRagon Institute of MGH, MIT and Harvard, Charlestown, Massachusetts, USA
cAIDS and Cancer Virus Program, Science Applications International Corporation-Frederick, National Cancer Institute, Frederick, Maryland, USA.
Received 28 August, 2009
Revised 1 April, 2010
Accepted 10 April, 2010
Correspondence to Marcus Altfeld, MD, PhD, Associate Professor, Harvard Medical School, Ragon Institute of MGH, MIT and Harvard, 149 13th Street, 6th floor, Charlestown, MA 02129, USA. E-mail: firstname.lastname@example.org
Background: Chronic HIV-1 infection is characterized by high levels of persistent immune activation. Both HIV-1-encoded Toll-like receptor 7/8 (TLR7/8) ligands and TLR ligands encoded by products of microbial translocation have been implicated in inducing and sustaining immune activation in infected individuals, but the consequences of simultaneous exposure to different TLR ligands are not well understood.
Objective: To examine the impact of preexposure of monocytes to HIV-1-encoded TLR8 ligands on their ability to respond to subsequent stimulation with microbial TLR2/4 ligands.
Method: Stable monocytic cell lines (THP-1-Blue-CD14 cells) or primary monocytes were stimulated with ligands for TLR2, TLR4, and TLR8, including chemically inactivated HIV-1, alone, or in sequential combinations. Responses by THP-1 cells to TLR stimulation were quantified using Quanti-Blue colometric assay, and TLR-induced tumor necrosis factor-α production of primary monocytes was quantified by intracellular cytokine staining using flow cytometry.
Results: The exposure of monocytes to HIV-1 or HIV-1-derived TLR8 ligands sensitized these cells for TLR4 stimulation, resulting in a significantly higher response to lipopolysaccharide compared to cells that were not prestimulated with TLR8 ligands or HIV-1.
Conclusion: TLR crosstalk can enhance the pro-inflammatory monocytes response to products of microbial translocation and might play an important role in the modulation of immune function in HIV-1 infection.
Chronic HIV-1 infection is characterized by strong persistent immune activation, and the level of immune activation has been identified as a significant predictor of HIV-1-disease progression [1,2]. One widely accepted model of HIV-1 immunopathogenesis postulates that heightened immune activation results in accelerated activation and proliferation of memory-effector CD4+T-cells, leading to their deletion . Recent studies investigating the mechanisms underlying HIV-1-associated immune activation have suggested that Toll-like receptors (TLRs) play a central role in mediating immune activation in HIV-1 infection [4–6].
TLRs recognize components of pathogens and initiate the pathogen-specific immune response [7,8]. TLR ligands have been shown to modulate monocyte function  initiating signaling cascades leading to nuclear factor-κB (NF-κB) activation and subsequent gene expression, which vary depending on which TLR has been activated . TLR2 and TLR4 are involved in the recognition of microbial products [11,12], whereas TLR7 and TLR8 sense single-stranded RNA (ssRNA) [7,13–15]. It has been demonstrated that chronic HIV-1 infection is associated with elevated circulating levels of the TLR4 ligand lipopolysaccharide (LPS), an important component of Gram-negative bacteria that enter the circulation through microbial translocation in the immunocompromised gut of HIV-1-infected individuals . Furthermore, the ssRNA sequence of HIV-1 itself contains sequences that can serve as TLR7 and TLR8 ligands and directly stimulate pro-inflammatory cytokine production by dendritic cells and monocytes [13,14,17,18]. Chronic stimulation of TLRs by products of microbial translocation and HIV-1-derived TLR ligands might, therefore, represent an important mechanism of immune activation in HIV-1 infection.
The consequences of stimulation through different TLR pathways for monocyte function are, however, not well understood. Recently, published studies on the murine model and human studies suggest that the activation of one TLR pathway in a cell has a significant impact on the ability of that cell to respond to stimulation of a different TLR pathway [19–21], resulting in either enhancement or reduction in the response to a second TLR ligand . The current study examined the impact of preexposure of monocytes to HIV-1 virions, the genomic RNA of which contains TLR ligands, on their ability to respond to subsequent stimulation with microbial TLR2/4 ligands, and demonstrates significant enhancement of the monocytes response to LPS following prestimulation with HIV-1.
Materials and methods
Samples from HIV-1-negative individuals enrolled at Massachusetts General Hospital were included in the studies. The study was approved by the Massachusetts General Hospital Institutional Review Board and the University of KwaZulu-Natal Biomedical Research Ethics Committee and each participant gave informed consent for participation.
THP-1 cell culture
THP-1-Blue-CD14 (THP-1) cells are TLR-expressing human monocytic NF-κB-reporter cells expressing CD14 and various TLRs. Upon TLR stimulation, THP-1 cells activate transcription factors resulting in the secretion of embryonic alkaline phosphatase (SEAP), which is detected using Quanti-Blue reagent, as described by the manufacturer's protocols (Invivogen, San Diego, California, USA). THP-1-Blue-CD14 cells were maintained in medium consisting of RPMI 1640 (Roswell Park Memorial Institute), 10% fetal calf serum (FCS), 2 mmol/l L-glutamine, supplemented with 200 μg/ml of zeocin and 10 μg/ml of blasticidin. THP-1 cells (3 × 105 cells/well) were stimulated in 96-well flat-bottomed plates with either medium alone or TLR ligands at the following concentrations: 2 μl/ml of heat-killed Listeria monocytogenes (HKLM, TLR2 ligand), 0.1 μg/ml of LPS (LPS-EK, TLR4 ligand), or 0.5 μg/ml CL075 (thiazoloquinolone derivative, TLR8 ligand) all from Invivogen  at 37°C and 5% CO2 for 18 h. The concentrations for TLR ligands and incubation periods were optimized in previous experiments (data not shown). TLR stimulation resulted in the activation of the NF-κB pathway and the NF-κB-dependent expression and secretion of the secreted alkaline phosphatase (SEAP) that was subsequently quantified using a Quanti-Blue mix assay and read at 630 nm according to the manufacturer's protocol (Invivogen).
In addition to stimulation with single TLR agonists, the effects of prestimulation or ‘priming’ of THP-1 cells with one TLR ligand (TLR8, or TLR2, or TLR4) on the response to stimulation with a second TLR ligand were assessed. THP-I cells were prestimulated with 0.5 μg/ml of CL075 or media alone in four separate wells. Supernatants from the THP-I-cell culture were removed after 3 h of incubation, the adhered cells washed twice with fresh RPMI, replenished with fresh media, and cultured with TLR2-HKLM, TLR4-LPS-EK, TLR8-CL075 (at similar concentrations indicated above), or media alone for a further 18 h, before NF-κB activation was quantified using the Quanti-Blue assay. In a subset of experiments, the TLR8 antagonist Pepinyl-MYD (Invivogen) was added to THP-1 cells (3 × 105 cells/well) for 6 h prior to stimulation with CL075 to demonstrate that the prestimulation with CL075 was indeed mediated through TLR8.
In-vitro stimulation of peripheral blood mononuclear cells with Toll-like receptor ligands
Fresh peripheral blood mononuclear cells (PBMCs) from healthy individuals were separated from whole blood by Ficoll-Hypaque (Sigma, St Louis, Missouri, USA). 1.5 × 106 PBMCs/ml in RPMI were cultured with at 2 μl/ml of TLR2-HKLM, 0.1 μg/ml of TLR4-LPS-EK, or 0.5 μg/ml of TLR8-CL075. The effect of prestimulation of PBMCs was assessed using the TLR8 CL075 agonist and also using aldrithiol-2-inactivated HIV-1 virus or microvesicle controls (MN strain, lot P4097; AIDS and Cancer Virus Program, National Cancer Institute). Fresh PBMCs were prestimulated with 0.5 μg/ml of CL075, 0.7 μg/ml of p24 (CA) aldrithiol-2 inactivated virus (AT-2 virus), or vesicle control, or media alone. The cultured PBMCs were then washed twice with RPMI after 3 h of prestimulation, and supernatants from the PBMC culture were removed and replenished with fresh media. The prestimulated PBMCs were then cultured with TLR2-HKLM, TLR4-LPS-EK, TLR8-CL075, AT-2 virus, vesicle control, or media alone (at the concentrations indicated above) for an additional 18 h. PBMCs that were preincubated with media, but no TLR agonists, acted as controls. Supernatants were collected for tumor necrosis factor-α (TNF-α) ELISA quantification according to the manufacturer's instructions (SA biosciences, Frederick, MD, USA). For the detection of intracellular cytokines using flow cytometric analysis, 5 μg/ml brefeldin A (Sigma) was added to each tube after adding the TLR ligands. Stimulation for all assays was conducted at 37°C and 5% CO2.
The intracellular cytokine content of monocytes was determined following 18 h of stimulation. Briefly, cells were stained for surface [CD3-Alexa700, CD19-Alexa700, CD56-Alexa700, CD14-allophycocyanin (APC)-Cy7] and intracellular marker TNF-PE-Cy7 (Becton Dickson, Franklin Lakes, New Jersey, USA). Monocytes were defined as CD3neg CD19neg CD56neg CD14+cells, and gates were set accordingly. Cells were fixed, permeabilized (Fix perm A&B; Caltag, Burlingame, California, USA), and stained intracellularly with anti-TNF-α-PE-Cy7. All samples were acquired on an LSRII (BD). The frequencies of cytokine-positive monocytes were determined by subsequent analysis using FlowJo software.
Statistical analyses and graphical presentation were done using Graphpad Prism 5 (Graphpad). Results are given as means with SDs or medians with ranges. Paired two-tailed Student's t-tests were used to test the statistical significance. Differences after comparisons were considered statistically significant if P value was less than 0.05.
Preexposure of THP-1-CD14 cells to Toll-like receptor 8 ligands modulates their responses to Toll-like receptor 2 and Toll-like receptor 4 ligands
We initially stimulated THP-1-cells with either medium alone or different TLR ligands (TLR2-HKLM, TLR4-LPS-EK, or TLR8-CL075) to determine their ability to respond to these ligands. Upon stimulation with these TLR2/4 and 8 ligands, THP-1-cells secreted considerable amount of SEAP, which was significantly detectable above background (TLR2; P ≤ 0.0001, TLR4; P = 0.001, TLR8; P ≤ 0.0001) using Quanti-Blue assay (Fig. 1a). As previous studies have suggested that the sequence of exposure to TLR agonists may impact on the ability of cells to respond to subsequent stimulation , we subsequently studied the effects of preexposure of THP-1-cells to a TLR8 agonist (agonist 1) on responses to TLR2 and TLR4 agonists (agonist 2; Fig. 1b). The prestimulation effects on the THP-1-cells were also assessed with a TLR2 or TLR4 ligand as the prestimulant agonist (agonist 1; Fig. 1c and d). No significant differences were noted in the response to TLR4 or TLR8 ligands when cells were prestimulated with the TLR2 agonist (Fig. 1c). TLR4 prestimulation significantly increased the response to the TLR8 ligand CL075 (P = 0.03), resulting in a two-fold increase compared to preincubation with media alone (Fig. 1d). However, the preexposure to TLR8 ligands modulated the ability of monocytes to respond to subsequent stimulation most dramatically, significantly increasing their responses to subsequent stimulation with TLR2 (three-fold, P = 0.02) and TLR4 (five-fold, P = 0.0002) ligands (Fig. 1b). In contrast, TLR8 ligands elicited no notable priming effects to stimulation with the same TLR8 agonist (P = 0.19). This priming effect was significantly higher than the sum of TLR4 and TLR8 stimulation, suggesting that the effect was not simply additive, but that TLR8 prestimulation amplified the response to TLR4 ligands. Using the TLR8 antagonist Pepinyl-MYD (Invivogen), we furthermore demonstrate that addition of this antagonist to THP-1 cells before exposing them to the TLR8 agonist significantly (P = 0.0001) reduced the subsequent response to LPS [average optical density of 0.52 ± 0.4 (SD)], compared to THP-1 cells that were prestimulated with the TLR8 agonist in the absence of the antagonist prior to stimulation with LPS [average optical density of 1.8 ± 0.2 (SD)]. In contrast, the antagonist did not decrease the response of THP-1 cells to LPS in the absence of prestimulation with the TLR8 agonist (data not shown). Taken together, these data using a monocyte-derived cell line demonstrate that preexposure to TLR8 ligands can significantly enhance the subsequent response of monocytes to TLR2 and TLR4 ligands.
Preexposure to Toll-like receptor 8 ligands increases the cytokine response of primary monocytes to Toll-like receptor 2 and Toll-like receptor 4 ligands
We next extended these initial studies to assess the effects of preexposure to TLR8 agonist on the subsequent response to TLR2 and TLR4 ligands by primary monocytes. Multiparameter flow cytometric analysis was employed to characterize the cytokine production by primary monocytes using intracellular cytokine staining. Representative flow cytometry plots showed robust TNF-α production by monocytes after stimulation with TLR2, TLR4, and TLR8 ligands alone compared to stimulation with medium alone (Fig. 2a). The effect of prestimulation of PBMCs with the TLR8 agonist CL075 was subsequently assessed. TLR8 ‘primed’ PBMCs were washed and then cultured with TLR2-HKLM, TLR4-LPS-EK, TLR8-CL075, or media alone. PBMCs precultured with medium alone were used as controls. The percentage TNF-α+ monocytes in response to the respective TLR agonists following prestimulation with medium alone or the TLR8 agonist CL075 revealed that preexposure to TLR8 agonists significantly (P = 0.02) increased monocyte responses to TLR4, but not TLR2 ligand (Fig. 2a and b). Once again, this priming effect was significantly higher than the sum of TLR4 and TLR8 stimulation alone, demonstrating that TLR8 prestimulation was amplifying responses to TLR4 ligands. In contrast, the percentage TNF-α+ monocytes following prestimulation with TLR8 agonist CL075 followed by re-stimulation with the TLR8 agonist revealed a decreased monocyte response. In conclusion, robust TNF-α production by monocytes is observed after stimulation with TLR2, TLR4, and TLR8 ligands, and responses to TLR4, but not TLR2 ligands, are significantly increased by preexposure to TLR8 agonists.
Preexposure to AT-2 virus increases the cytokine response of primary monocytes to the Toll-like receptor 4 ligand lipopolysaccharide
To further assess the impact of preexposure to HIV-1-derived TLR ligands on the subsequent response of the primary human monocytes to microbial LPS, we stimulated monocytes with AT-2-inactivated HIV-1, as described previously , and observed robust TNF-α production compared to vesicle and medium control (Fig. 2c and d). Similar to TLR8 ligands, prestimulation with the AT-2 virus significantly enhanced the subsequent response of monocytes to LPS compared to cells that had preexposure to medium alone (P = 0.02) or vesicle control (P = 0.03) prior to stimulation with LPS. Furthermore, the response of monocytes to LPS following prestimulation with AT-2 virus was also significantly (two-fold) higher than the sum of the responses to LPS and AT-2 virus alone (P = 0.035; Fig. 2d). In line with these data, prestimulation with the TLR8 ligands significantly (P = 0.03) increased the subsequent amount of secreted TNF-α produced in response to LPS stimulation quantified by ELISAs in the culture supernatants [average of 15 pg TNF-α/ml ±1.1 (SD) in the absence of prestimulation versus 58 pg TNF-α/ml ±2.8 (SD) following prestimulation with the TLR8 ligand CL075, P < 0.03, data not shown]. In conclusion, these data demonstrate that preexposure to HIV-1 significantly enhances the response of monocytes to microbial TLR4 ligands such as LPS.
Stimulations of the TLR7/8 pathway by HIV-1-ssRNA-derived ligands and the TLR4 pathways by products for microbial translocation have been both implicated in mediating the activation of the immune system during chronic HIV-1 infection, and in resulting immunopathogenesis [16,18]. However, little is known about how these two pathways might influence each other. Here we demonstrate that exposure to HIV-1 or HIV-1-derived TLR8 ligands significantly increases the production of pro-inflammatory cytokines by monocytes in response to subsequent stimulation with the TLR4 ligand LPS. These data suggest that the combination of HIV-1 replication and microbial translocation in chronically HIV-1-infected individuals might not only additively increase the activation of the immune system, but might actually amplify these effects.
The observation that stimulation through TLR8 amplifies the pro-inflammatory cytokine response of monocytes in response to products of microbial translocation, such as LPS, has important implications for HIV-1 pathogenesis, as the levels of immune activation during chronic HIV-1-infection have been strongly associated with the loss of CD4+ T cells and consecutive disease progression . The increased sensitivity of monocytes to LPS in the setting of TLR8 stimulation also helps to explain the observation that HIV-1-associated immune activation decreases rapidly and significantly following the initiation of antiretroviral therapy and resulting reduction of viral load [25–27], even before reconstitution of the immune system and eventual reduction of microbial translocation , as the decrease in the levels of HIV-1-encoded TLR8 ligands might render monocytes less responsive to circulating levels of LPS. Taken together, these data support a model in which TLR crosstalk plays an important role in the modulation of immune function in HIV-1 infection, and in the ability of the infected host to respond to stimulation by a variety of TLR ligands it is exposed to, either through intestinal microbial translocation, other opportunistic pathogens, or HIV-1 itself.
These studies were supported by the Howard Hughes Medical Institute through the Kwa-Zulu Natal Research Institute for TB and HIV and the Doris Duke Charitable Foundation and supported in part with federal funds from the National Cancer Institute, National Institutes of Health, under contract HHSN266200400088C.
M.W.M. performed the experiments and wrote the article. J.J.C. helped with the development, planning, and optimization of the assays. J.D.L. provided the AT-2 virus. T.N. helped with the design of the studies and the preparation of the article. M.A. designed the studies and wrote the article.
The present study is presented in 5th IAS meeting, Cape Town, 2009.
1. Deeks SG, Walker BD. The immune response to AIDS virus infection: good, bad, or both? J Clin Invest 2004; 113:808–810.
2. Giorgi JV, Fahey JL, Smith DC, Hultin LE, Cheng HL, Mitsuyasu RT, Detels R. Early effects of HIV on CD4 lymphocytes in vivo. J Immunol 1987; 138:3725–3730.
3. Fahey JL, Taylor JM, Manna B, Nishanian P, Aziz N, Giorgi JV, Detels R. Prognostic significance of plasma markers of immune activation, HIV viral load and CD4 T-cell measurements. AIDS 1998; 12:1581–1590.
4. Chang JJ, Altfeld M. TLR-mediated immune activation in HIV. Blood 2009; 113:269–270.
5. Douek D. HIV disease progression: immune activation, microbes, and a leaky gut. Top HIV Med 2007; 15:114–117.
6. Haynes BF. Gut microbes out of control in HIV infection. Nat Med 2006; 12:1351–1352.
7. Akira S, Hemmi H. Recognition of pathogen-associated molecular patterns by TLR family. Immunol Lett 2003; 85:85–95.
8. Takeda K, Akira S. TLR signaling pathways. Semin Immunol 2004; 16:3–9.
9. Bekeredjian-Ding I, Roth SI, Gilles S, Giese T, Ablasser A, Hornung V, et al. T cell-independent, TLR-induced IL-12p70 production in primary human monocytes. J Immunol 2006; 176:7438–7446.
10. Takeda K, Akira S. Toll-like receptors in innate immunity. Int Immunol 2005; 17:1–14.
11. Akamine M, Higa F, Arakaki N, Kawakami K, Takeda K, Akira S, Saito A. Differential roles of Toll-like receptors 2 and 4 in in vitro responses of macrophages to Legionella pneumophila. Infect Immun 2005; 73:352–361.
12. Poltorak A, He X, Smirnova I, Liu MY, Van Huffel C, Du X, et al. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 1998; 282:2085–2088.
13. Heil F, Hemmi H, Hochrein H, Ampenberger F, Kirschning C, Akira S, et al. Species-specific recognition of single-stranded RNA via Toll-like receptor 7 and 8. Science 2004; 303:1526–1529.
14. Meier A, Alter G, Frahm N, Sidhu H, Li B, Bagchi A, et al. MyD88-dependent immune activation mediated by human immunodeficiency virus type 1-encoded Toll-like receptor ligands. J Virol 2007; 81:8180–8191.
15. Hornung V, Ellegast J, Kim S, Brzozka K, Jung A, Kato H, et al. 5′-Triphosphate RNA is the ligand for RIG-I. Science 2006; 314:994–997.
16. Brenchley JM, Price DA, Schacker TW, Asher TE, Silvestri G, Rao S, et al. Microbial translocation is a cause of systemic immune activation in chronic HIV infection. Nat Med 2006; 12:1365–1371.
17. Beignon AS, McKenna K, Skoberne M, Manches O, DaSilva I, Kavanagh DG, et al. Endocytosis of HIV-1 activates plasmacytoid dendritic cells via Toll-like receptor-viral RNA interactions. J Clin Invest 2005; 115:3265–3275.
18. Meier A, Chang JJ, Chan ES, Pollard RB, Sidhu HK, Kulkarni S, et al. Sex differences in the Toll-like receptor-mediated response of plasmacytoid dendritic cells to HIV-1. Nat Med 2009; 15:955–959.
19. Bagchi A, Herrup EA, Warren HS, Trigilio J, Shin HS, Valentine C, Hellman J. MyD88-dependent and MyD88-independent pathways in synergy, priming, and tolerance between TLR agonists. J Immunol 2007; 178:1164–1171.
20. Napolitani G, Rinaldi A, Bertoni F, Sallusto F, Lanzavecchia A. Selected Toll-like receptor agonist combinations synergistically trigger a T helper type 1-polarizing program in dendritic cells. Nat Immunol 2005; 6:769–776.
21. Sato S, Nomura F, Kawai T, Takeuchi O, Muhlradt PF, Takeda K, Akira S. Synergy and cross-tolerance between Toll-like receptor (TLR) 2- and TLR4-mediated signaling pathways. J Immunol 2000; 165:7096–7101.
22. Lester RT, Yao X-D, Blake BT, McKinnon LR, Kaul R, Wachihi C, et al. Toll-like receptor expression and responsiveness are increased in viraemic HIV-1 infection. AIDS 2008; 22:685–694.
23. Gorden KB, Gorski KS, Gibson S, Kedl R, Kieper W, Qiu X, et al. Synthetic TLR agonists reveal functional differences between human TLR7 and TLR8. J Immunol 2005; 174:1259–1269.
24. Lee PI, Ciccone EJ, Read SW, Asher A, Pitts R, Douek DC, et al. Evidence for translocation of microbial products in patients with idiopathic CD4 lymphocytopenia. J Infect Dis 2009; 199:1664–1670.
25. Giorgi JV, Landay A. HIV infection: diagnosis and disease progression evaluation. Methods Cell Biol 1994; 42(Pt B):437–455.
26. Gray CM, Schapiro JM, Winters MA, Merigan TC. Changes in CD4+ and CD8+ T cell subsets in response to highly active antiretroviral therapy in HIV type 1-infected patients with prior protease inhibitor experience. AIDS Res Hum Retroviruses 1998; 14:561–569.
27. Lempicki RA, Kovacs JA, Baseler MW, Adelsberger JW, Dewar RL, Natarajan V, et al. Impact of HIV-1 infection and highly active antiretroviral therapy on the kinetics of CD4+ and CD8+ T cell turnover in HIV-infected patients. Proc Natl Acad Sci U S A 2000; 97:13778–13783.
28. Brenchley JM, Schacker TW, Ruff LE, Price DA, Taylor JH, Beilman GJ, et al. CD4+ T cell depletion during all stages of HIV disease occurs predominantly in the gastrointestinal tract. J Exp Med 2004; 200:749–759.
This article has been cited 3 time(s).
AIDS Research and Human RetrovirusesA Role for TLR Signaling During B Cell Activation in Antiretroviral-Treated HIV IndividualsAIDS Research and Human Retroviruses
Viruses-BaselLearning from the Messengers: Innate Sensing of Viruses and Cytokine Regulation of Immunity-Clues for Treatments and VaccinesViruses-Basel
Fish & Shellfish ImmunologyMolecular characterization and expression analysis of TLR 7 and TLR 8 homologs in large yellow croaker (Pseudosciaena crocea)Fish & Shellfish Immunology
HIV-1; lipopolysaccharide; microbial translocation; monocytes; Toll-like receptors
© 2010 Lippincott Williams & Wilkins, Inc.
Highlight selected keywords in the article text.