Inhibitory Effects of Rosmarinic Acid on Lck SH2 Domain Binding to a Synthetic Phosphopeptide

Soon Cheol Ahn1, 2, Won Keun Oh1, Bo Yeon Kim1, Dae Ook Kang1, 3, Min Soo Kim1, Gun Young Heo1, Jong Seog Ahn1

Abstract

In the course of screening inhibitors from the methanol (MeOH) extracts of 168 medicinal plants against lymphocyte cell-specific kinase (Lck) Src-homology 2 (SH2) binding to a synthetic phosphotyrosine-containing peptide (phosphopeptide), we isolated rosmarinic acid from the MeOH extract of Prunella vulgaris, which showed specific inhibitory activity. The IC₅₀ value for Lck SH2 binding to phosphopeptide (SGSGEEPQpYEEIPI) of hamster polyomavirus middle-sized tumor (hmT pY324) was 7 μM. However, even at concentrations of 0.1 to 1000 μM, no significant inhibitions were observed against other SH2 domains binding such as the growth factor receptor binding protein 2 (Grb2) SH2 domain to phosphopeptide of Shc and phospholipase Cγ1 (PLCγ1) SH2 domain to translational elongation factor 1α (EF1α) C-terminal. Rosmarinic acid inhibited interleukin-2 (IL-2) gene expression by 50 % at a concentration of 8 μM in Jurkat cells stimulated with anti-CD3 and anti-CD4 antibodies. FK506 and cyclosporin A (CsA) employed as positive controls showed less than 30 % inhibition at the same concentration. In addition, rosmarinic acid inhibited the intracellular [Ca²⁺]_i increase in Jurkat cells after T cell activation in a dose-dependent manner at concentrations of 1.4 to 140 μM of rosmarinic acid, which is one of the earliest responses of antigen-specific T cell receptor (TCR) and of the upstream pathway of IL-2 expression. Taken together, these results suggest that rosmarinic acid has the potential to specifically inhibit Lck SH2 domain binding to its cognate ligand, including ZAP-70, Cbl, HS-1, and PLCγ1, and Lck-dependent Ca²⁺ signaling pathway of its downstream effector and finally to modulate IL-2 gene expression after T cell activation.

Key words

Prunella vulgaris - Labiatae - rosmarinic acid - Lck SH2 - Grb2 SH2 - PLC γ1 SH2 - IL-2 expression - intracellular [Ca²⁺]_i.

Introduction

Stimulation of the TCR initiates intracellular signaling events that lead to T cell differentiation and proliferation. Several pathways have been implicated in this activation process, notably the induction of tyrosine phosphorylation and the phosphatidylinositol (PI) and Ras signaling pathways. A number of protein tyrosine kinases has been shown to be important in T cell activation. One such kinase, p56^lck, is specifically expressed in T lymphocyte as one of the Src family of non-receptor protein tyrosine kinases. The protein contains one SH2 domain, which is a protein/protein interaction motif composed of approximately 100 amino acids and binds to phosphotyrosine (pY) and certain adjacent sequences [1]. The SH2 domain of Lck has been demonstrated to be genetically indispensable for the T cell activation [2]. A recent analysis of the crystal structure of the Lck SH2 domain suggested that it contains a high-affinity motif pY324 with the sequence EPQpYEEIPIYL [3]. The importance of the interaction and the high specificity of binding between Lck SH2 domain and a number of tyrosine-phosphorylated signaling proteins, including ZAP-70, Cbl, HS-1, and PLCγ1 have been shown [4]. Moreover, it has been suggested that the Lck SH2 domain might additionally contribute to TCR signal transduction as an adapter molecule for the recruitment of phosphoproteins to signaling complexes [5].

IL-2 gene expression is a common measure of T cell activation and requires the co-operation of several signaling pathways and transcription factors, including AP-1, NFAT, Oct-1, and NFκB [6]. These transcription factors bind to the IL-2 gene regulatory elements and induce transcription. Recent evidence suggests that the synthesis of IL-2 is mediated by an increase in intracellular [Ca²⁺]_i through phosphorylated PLCγ1, and that the control of Ca²⁺ homeostasis plays a critical role in the regulation of IL-2 synthesis [7]. The resulting increase in cytoplasmic [Ca²⁺]_i activates calcineurin, a Ca²⁺/calmodulin-dependent serine/threonine phosphatase, which mediates the translocation of transcription factor NFAT from the cytoplasm to the nucleus. Then NFAT binds directly to its proximal binding site of the IL-2 promoter and activates the T cell-specific transcription factors, which leads to the up-regulation of IL-2 gene transcription [8].

Recently significant attention has been focused on the development of potent and selective inhibitors of specific SH2 domains, as a result of considerations of the general impact on diseases, such as restenosis, cancers, cardiovascular disease, osteoporosis, and apoptosis [9]. Moreover, inhibitors of specific SH2 domain binding have become important therapeutic targets in the treatment and prevention of these diseases. In particular, considerable research has been performed on Src, PI-3-kinase, and Grb2 and more recently upon Lck. The Lck SH2 domain may serve as a potent target for specific inhibitors that block the T cell activation signal due to its highly specific binding property. We screened for natural products that may modulate T cell signaling pathways using an Lck SH2 binding assay and consequently isolated a specific inhibitor of Lck SH2 binding, rosmarinic acid, from a Korean medicinal plant P. vulgaris. In this paper, we present the effects of rosmarinic acid on T cell activation prior to the development and evaluation of its use versus potential future targets.

Materials and Methods

Plant material

The herb of P. vulgaris was collected at Gaeryong Mountain, Taejon, Korea in April 1997 and identified by Dr. H. S. Lee of KRIBB. A voucher specimen (PB008 - 017) was deposited in the Korea Plant Extract Bank, Korea Research Institute of Bioscience and Biotechnology, Taejon, Korea.

Extraction and isolation

The air-dried whole herbs of P. vulgaris (600 g) were extracted three times with MeOH at room temperature and the extract was concentrated under reduced pressure to yield 18.7 g of an extract. The extract was dissolved in water and the suspension was extracted with ethyl acetate (EtOAc). This extract was subjected to silica gel (200 g, 70 - 230 mesh, Merck Co., Germany) column chromatography eluting with CHCl₃ : MeOH : H₂O (10 : 5 : 1, 5 L). The eluate was collected every 500 mL. Fractions were monitored by TLC on silica gel plates (0.2 mm thickness, Merck Co., Darmstadt, Germany) with ethyl acetate : formic acid :water (44 : 3 : 3) and inhibition of Lck SH2 binding. The inhibitory fractions were collected and further purified through Sephadex LH-20 (φ2 × 90 cm, Sigma Co., St. Louis, MO, USA) using MeOH and the inhibitory pools were applied to HPLC (YMC-Gel C18, φ2 × 15 cm, YMC Co., Kyoto, Japan) using MeOH:0.5 % phosphoric acid (45 : 55) at a flow rate of 4 mL/min, followed by Diaion HP-20 (Nippon Rensui Co., Tokyo, Japan) adsorption, yielding 50 mg of rosmarinic acid.

Lck SH2 domain binding assay

The Lck SH2 binding assay was performed as previously described with minor modification [10]. The GST-Lck SH2 domain was diluted to 1 μM in PBS and a 100 μL aliquot was placed in each well of a maleic anhydride-precoated 96-well microplate (Pierce Chemical Co., Rockford, IL, USA). After 2 h incubation, the plates were washed three times with PBST (0.05 % Tween-20 in PBS). One hundred μL of 1 μM synthetic biotinylated phosphotyrosyl peptide (SGSGEEPQpYEEIPI) derived from the sequence of hamster polyomavirus middle-sized tumor (hmT pY324) was then mixed with 10 μL of test sample and the mixture was placed on the coated protein for 1 h. After washing twice with PBST, horseradish peroxidase (HRP)-conjugated streptavidin (Pierce Chemical Co.) was added to each well to detect the bound biotinylated phosphopeptide. HRP activity was determined by 3,3′,5,5′-tetramethylbenzidine (TMB) oxidation and optical density was observed on a plate reader at 450 nm.

Grb2 SH2 domain binding assay

The Grb2 SH2 binding assay was performed as previously described with minor modification [11]. Protein A-coated scintillation proximity assay (SPA) beads (Amersham Co., Arlington Heights, IL, USA) and a rabbit IgG anti-GST antibody (Molecular Probes, Eugene, OR, USA) were used to detect binding affinity. The reaction mixture in each well of a 96-well microplate contained 0.25 μg GST-Grb2 SH2 domain fusion protein, 6 μg of anti-GST antibody, 0.24 mg of protein A SPA beads, and 0.2 μCi [³H]propionyl-labeled phosphopeptide (SpYVNVK) which was derived from the sequence of the human Shc (Shc pY317). The final volume was adjusted to 200 μL by adding a buffer containing 20 mM Tris-HCl (pH 7.4), 250 mM NaCl and 0.1 % bovine serum albumin. The plates were incubated with shaking at room temperature for 3 h. The binding activity was determined by measuring the energy released from SPA bead-bound [³H]phosphopeptide on a 1450 Wallac Microbeta counter (PE, Foster City, CA, USA).

PLCγ1 SH2 domain binding assay

The PLCγ1 SH2 binding assay was performed as previously described with minor modification [12]. The fragment of PLCγ1 containing the SH2,SH2,SH3 domain (54 KDa) was diluted to 0.4 μM in PBS and a 50 μL aliquot was coated onto an Immuno-Maxisorp microwell plate (Nunc Co., Naperville, IL, USA) for 4 h at room temperature. Fifty μL of 0.2 μM GST-EF1α C-terminal fragment (186 amino acids) was then mixed with 10 μL of test sample, and the mixture was placed on the coated protein for 4 h. Anti-GST antibody was added to each well and the well plate was allowed to stand for 4 h. Goat anti-mouse IgG, IgM and IgA antibody-conjugated HRP were then added to detect the bound GST antibody. HRP activity was determined by O-phenylenediamine (OPD) oxidation and optical density was observed on a plate reader at 492 nm.

IL-2 transcriptional expression

Five × 10⁶ Jurkat T cells were transfected with 5 μg of NFAT IL-2 luciferase plasmid DNA [13] using Superfect (Qiagen GmbH, Hilden, Germany). The cells were incubated for 24 h and then treated with various concentrations of rosmarinic acid, CsA (Calbiochem, San Diego, CA, USA) and FK506 (Calbiochem, San Diego, CA, USA) as immunosuppressants and LY294002 (Calbiochem) as a PI-3-K inhibitor for 30 min prior to T cell activation by immobilized anti-CD3 and anti-CD4 antibodies made in our laboratory. Fourteen hours later, cells were harvested, washed twice with PBS and then lysed in reporter lysis buffer (Promega Corp., Madison, WI, USA). Cellular debris was removed by centrifugation and cell extracts were assayed for luciferase activity using a Berthold luminometer LB953 (PE).

Intracellular [Ca²⁺]_i increase

Jurkat T lymphocytes were cultivated in RPMI 1640 medium (GifcoBRL, Rockville, MD, USA) supplemented with 10 % (v/v) fetal bovine serum (Hyclone, Logan, UT, USA). The cells were suspended in 5 mL of medium containing 125 mM NaCl, 5.9 mM KCl, 1.2 mM MgCl₂, 1.3 mM CaCl₂, 3 mM glucose, and 25 mM HEPES (pH 7.4). Fluorescent calcium-binding dye, Fura-2 acetoxymethyl ester (Fura-2/AM) was then added and the cells were allowed to accumulate Fura-2/AM for 40 min at 37 °C in the dark. After loading, the cells were spun down and thoroughly washed. Various concentrations of rosmarinic acid were treated for 10 min and the cells were then transferred to a cuvette placed in a spectrophotofluorometer. Fluorescence intensities (Δ) were measured before and after stimulation with saturating concentrations of anti-CD3 and anti-CD4 antibodies and subsequent cross-linking with the appropriate secondary antibody.

Statistical analysis

The statistical significance of differences between groups was evaluated for two parallel experiments using the Student’s t-test. The experiments were performed in duplicate and the standard deviations were indicated as bars.

Results

Isolation of rosmarinic acid

For the purpose of screening inhibitors from the MeOH extracts of 168 medicinal plants against Lck SH2 binding to phosphopeptide, we have employed a modified ELISA system based on the principle that GST fusion protein of Lck SH2 domain binds to its cognate phosphotyrosine (pY)-containing peptides [10]. As the result of screening, P. vulgaris among MeOH extracts showed the highest inhibitory activity and was chosen as a candidate for inhibitor. Finally an inhibitor was isolated in 95 % purity, yielding 50 mg of an off-white powder, by activity-guided purification procedures from the MeOH extract of P. vulgaris by performing EtOAc extraction, and silica gel, Sephadex LH-20, Lobar, and Diaion HP-20 column chromatography and HPLC. The inhibitor was then identified as rosmarinic acid (Fig. [1]) by comparing its ¹H-NMR and ¹³C-NMR data with those reported [14] and by comparing its Lck SH2 binding inhibitory activity and HPLC retention time versus authentic rosmarinic acid (Indofine Chemical Co., Belle Mead, NJ, USA) (data not shown). The rosmarinic acid was the (R)-form as shown by its specific optical rotation of 106° [15]. Also the (S)-form of rosmarinic acid was synthetically obtained [16] and its inhibitory activity was shown to be same as the purified rosmarinic acid from P. vulgaris.

Fig. 1 Structure of rosmarinic acid.

Effect of rosmarinic acid on the Lck SH2 domain binding

To determine the effect of rosmarinic acid on the binding of the Lck SH2 domain to synthetic phosphopeptide (SGSGEEPQpYEEIPI, hmT pY324), rosmarinic acid was mixed and treated with phosphopeptide at concentrations of 0.1 to 1000 μM. As a result, its inhibitory activity against Lck SH2 domain binding was examined in a dose-dependent manner and its IC₅₀ value determined to be 7 μM (Fig. [2], •). Interestingly, when phosphopeptide was added after pretreating the coated GST-Lck SH2 fusion protein with rosmarinic acid for 1 h and this was followed by washing, rosmarinic acid was found to inhibit Lck SH2 binding in the same manner as above described but its IC₅₀ value was decreased to 0.6 μM (Fig. [2], ○). This result suggests that rosmarinic acid inhibits Lck SH2 domain binding, and that its inhibitory activity is due to competitive blocking of the binding site of the synthetic phosphotyrosine peptide on the Lck SH2 domain, and is not due to the disruption of assay conditions caused by, for example, the coating protein in the preparation of reagents.

Fig. 2 Inhibitory activity of rosmarinic acid on the SH2 binding. Rosmarinic acid was treated at concentrations of 0.1 to 1000 μM to determine the GST-Lck SH2 binding to biotinylated phosphotyrosyl peptide (-SGSGEEPQpYEEIPI, pY324) of hmT when phosphopeptide was added after pretreatment of rosmarinic acid (○), or together with rosmarinic acid (•) on the coated GST-Lck SH2. Also the PLCγ1 SH2 binding to GST-EF1α C-terminal (▴) and the GST-Grb2 SH2 binding to propionyl-labeled phosphotyrosyl peptide (SpYVNVK, pY317) of human Shc () were determined. The experiments were performed in duplicate and the standard deviations were indicated as bars.

Effects of rosmarinic acid on the SH2 domain binding of other species

It has been reported that the binding sequences of Lck SH2 domain to the target proteins differ from those of intracellular signaling proteins, such as PLCγ1, Src, PI-3 kinase, Ras-GAP or Grb2. To provide evidence to support the specificity of the inhibitory effects of rosmarinic acid on the binding of the Lck SH2 domain to its target phosphotyrosine-containing peptide, we examined the inhibitory effects of rosmarinic acid on other species SH2 domains binding, such as PLCγ1 to the C-terminal fragment of EF1α (Fig. [2], ▴), and Grb2 to the phosphopeptide (SpYVNVK, pY317) of human Shc (Fig. [2], ) at concentrations of 0.1 to 1000 μM. However, no significant inhibitory activities were observed against both PLCγ1 SH2 domain binding and Grb2 SH2 domain binding as shown in Fig. [2], and even at a concentration of 150 μM, their inhibitory activities were less than 30 %. Although different experimental conditions were used in the binding assay, great differences in IC₅₀ values were observed between the bindings of these SH2 domain-containing proteins. This means that rosmarinic acid does not show broad inhibition versus the SH2 domains of general target proteins, but rather shows specific inhibition against the SH2 domain binding of Lck.

Effect of rosmarinic acid on anti-CD3 and anti-CD4 antibodies induced-IL-2 gene expression in Jurkat T cells

As a first approach to determine the biological potential in T cell activation, we examined the effect of rosmarinic acid on the IL-2 gene expression in Jurkat T cells by measuring the IL-2 promoter-driven reporter gene activity induced by immobilized anti-CD3 and anti-CD4 antibodies. Jurkat T cells were transfected with NFAT IL-2 promoter-luciferase reporter gene plasmid and then luciferase activity was measured after treatment with its substrate and buffer. CsA and FK506, known immunosuppressants, were used as positive controls, and LY294002, a known PI-3-K inhibitor, was used as a negative control. Rosmarinic acid inhibited the NFAT IL-2 promoter activity induced by immobilized anti-CD3 and anti-CD4 antibodies, with an IC₅₀ of 8 μM without any cytotoxicity (Fig. [3]). CsA and FK506 also represented inhibition against the IL-2 transcriptional expression but the inhibitory activities of those were less than 30 % at the same concentration of 8 μM (Fig. [3]). No inhibitory effect was shown by LY294002. It seems that rosmarinic acid is a more potent inhibitor than CsA or FK506 at 8 μM.

Fig. 3 Effect of rosmarinic acid on the IL-2 gene expression in Jurkat T lymphocytes. Jurkat T cells were transfected with 5 μg of NFAT IL-2 promoter-luciferase reporter plasmid DNA. After 24 h, the cells were treated with blank chemical (control), rosmarinic acid (RA), cyclosporin A (CsA), FK506, or LY294002 at a concentration of 8 μM, respectively, prior to T cell activation by immobilized anti-CD3 and anti-CD4 antibodies. IL-2 gene expression was detected by the luciferase reporter assay. The experiments were performed in duplicate and the standard deviations were indicated as bars.

Effect of rosmarinic acid on anti-CD3 and anti-CD4 antibodies-induced intracellular [Ca²⁺]_i increase in Jurkat T cells

It is known that the intracellular [Ca²⁺]_i level is elevated shortly after the cross-linking with anti-CD3 and anti-CD4 antibodies and Lck is required in this process. As a second approach to investigate the effect of rosmarinic acid on the contribution of the Lck SH2 domain to T cell signaling, we examined intracellular [Ca²⁺]_i increases after T cell activation. After pretreatment with Ca²⁺-binding fluorescence dye, Fura-2/AM, Jurkat T cells were incubated with 1.4 to 140 μM of rosmarinic acid for 10 min, prior to TCR stimulation by cross-linking with anti-CD3 and anti-CD4 antibodies. As shown in Fig. [4], the intracellular [Ca²⁺]_i increase was inhibited by rosmarinic acid in a dose-dependent manner, and without any cytotoxicity.

Fig. 4 Effect of rosmarinic acid on the intracellular [Ca²⁺]_i increase in Jurkat T lymphocytes. Jurkat T cells were loaded with Fura-2/AM, thoroughly washed and treated with various concentrations of rosmarinic acid (RA) for 10 min (○: without RA, •: RA 1.4 μM, ▴: RA 2.8 μM, : RA 28 μM, ▾: RA 140 μM). The cells were placed in a cuvette and fluorescence levels were measured before and after stimulation with saturating concentrations of anti-CD3 and anti-CD4 antibodies and subsequent cross-linking with the appropriate secondary antibody.

Discussion and Conclusion

Numerous lines of evidence have indicated that SH2 domains bind to their tyrosine-phosphorylated protein targets with high specificity [4]. The binding specificity between the SH2 domain and its counterpart was determined by the sequence and the structural context of the phosphotyrosine. We employed a modified ELISA system, as previously described [10], to screen natural product inhibitors against Lck SH2 binding, binding of GST-Lck SH2 to its cognate phosphopeptides in the presence of candidate inhibitors in a 96-well microplate. Several positive extracts were selected and finally rosmarinic acid was isolated from the MeOH extract of P. vulgaris. Rosmarinic acid has previously been shown to have various biological effects, including antimicrobial, antioxidative, and anti-inflammatory effects [14][17]. However, the inhibitory potential of rosmarinic acid in T cell activation has, to the best of our knowledge, not been reported until now, and this appears to be the first report of a natural product, which inhibits the binding of the Lck SH2 domain to its ligand. Therefore rosmarinic acid may modulate some T cell activation pathways mediated by the SH2 domain of Lck protein tyrosine kinase.

The best characterized early signaling cellular responses in activated T cells are the tyrosine phosphorylation of target proteins and the activation of PLC, calcineurin and Ras, which lead to the transcriptional activation of the IL-2 gene and subsequent T cell proliferation [18]. In addition, recent results suggest that the synthesis of IL-2 is mediated by an increase in intracellular [Ca²⁺]_i through activated PLCγ1 [7]. This increase activates calcineurin, which mediates nuclear translocation and induces the T cell-specific transcriptional nuclear factor required for the activation of T cells [6]. Therefore, as a further approach, we investigated the effect of rosmarinic acid on the contribution of the Lck SH2 domain to T cell signaling. Rosmarinic acid inhibited NFAT IL-2 promoter activity and the intracellular [Ca²⁺]_i increase. The triggering of T cells via TCR increases intracellular reactive oxygen species levels, and moreover, antioxidants effectively down-regulate early activation events, such as the elevation of intracellular [Ca²⁺]_i and increased tyrosine phosphorylation [19]. Since rosmarinic acid has potent anti-oxidative activity, it might suppress IL-2 expression and intracellular [Ca²⁺]_i increase induced by scavenging reactive oxygen species in turn induced by TCR activation. Nevertheless, our results suggest that rosmarinic acid blocks, at least partially, some step of the signal transduction pathway between TCR stimulation and IL-2 gene expression by inhibiting the binding of the Lck SH2 domain to its counterpart and ultimately inhibiting the IL-2 expression induced by immobilized anti-CD3 and anti-CD4 antibodies during T cell activation. It was previously found that the increased [Ca²⁺]_i flux followed by TCR stimulation is defective in the Lck-deficient T cell line JCaM1 but is reconstituted when functional Lck is transfected in the JCaM1 cell line [20]. Together with our observation, this can be interpreted to mean that rosmarinic acid has the potential to inhibit Lck-dependent Ca²⁺ signaling, and that the inhibitory effects of rosmarinic acid on IL-2 expression and intracellular [Ca²⁺]_i increase, after T cell activation are dependent on the blocking of the functional interaction of the Lck SH2 domain with the ligands of downstream effector molecules by rosmarinic acid.

These results suggest that rosmarinic acid has the potential to inhibit Lck-dependent IL-2 gene expression and intracellular [Ca²⁺]_i flux after T cell activation by modulating the interaction between the Lck SH2 domain and its cognate ligand. The present study describes the inhibitory effects of rosmarinic acid on the interaction between the Lck SH2 domain and synthetic phosphopeptide in vitro, and its effects on IL-2 expression and intracellular [Ca²⁺]_i increase after T cell activation in Jurkat cells. Because the expression of Lck is restricted to T lymphocyte cells and the binding of the Lck SH2 domain is highly specific, we suggest that rosmarinic acid be reviewed as a candidate drug for the treatment of diseases caused by the abnormal regulation of T cells. Studies are under way to determine the nature of the inhibitory mechanism from TCR stimulation to IL-2 gene expression during the process of T cell activation.

Acknowledgements

The authors thank Dr. B. M. Kwon of KRIBB for help in the binding assay of Grb2 SH2 domain and Prof. S. H. Ryu of Pohang University of Technology for providing the materials in PLCγ1 SH2 binding assay. In addition, we express our thanks to Prof. Y. Yoon of Ewha Women’s University and Dr. J. W. Won of Mokam Biotechnology Research Institute in measurement of IL-2 gene expression and [Ca²⁺]_i flux.

References

Dr. Jong Seong Ahn
Korea Research Institute of Bioscience and Biotechnology
P.O. Box 115
Yusong
Taejon 305-600
Korea
Phone: +82-42-860-4312
Fax: +82-42-860-4595
Email: jsahn@kribb.re.kr