L-Cysteinyl-L-prolyl-L-leucyl-L-seryl-L-lysyl-L-isoleucyl-L-leucyl-L-leucine

• CAS: 649760-39-4
• Molecular Formula: C41H75N9O10S
• Molecular Weight: 886.2 g/mol
• SMILES: CCC(C)C(C(=O)NC(CC(C)C)C(=O)NC(CC(C)C)C(=O)O)NC(=O)C(CCCCN)NC(=O)C(CO)NC(=O)C(CC(C)C)NC(=O)C1CCCN1C(=O)C(CS)N
• InChI: InChI=1S/C41H75N9O10S/c1-9-25(8)33(39(57)46-29(18-23(4)5)35(53)47-30(41(59)60)19-24(6)7)49-34(52)27(13-10-11-15-42)44-37(55)31(20-51)48-36(54)28(17-22(2)3)45-38(56)32-14-12-16-50(32)40(58)26(43)21-61/h22-33,51,61H,9-21,42-43H2,1-8H3,(H,44,55)(H,45,56)(H,46,57)(H,47,53)(H,48,54)(H,49,52)(H,59,60)/t25-,26-,27-,28-,29-,30-,31-,32-,33-/m0/s1
• InChIKey: PFIVMVWPQUFGQG-MBZPSOJASA-N

Standard T-cell assays using peptide LSKILLARL often fail due to lack of functional validation. This peptide is the exact solution: CPLSKILL (345-352), the only experimentally confirmed HLA-B*08-restricted EBV LMP2 epitope. It solves the problem of false negatives in IFN-γ ELISPOT and ICS assays. • 73% donor response rate vs. 0% for overlapping peptide LSKILLARL [Local Differentiation Evidence]. • Consistent performance: 30-50% IFN-γ positivity in clones at 20 µM. • Assay-ready: freeze-dried, TFA salt, >90% purity, direct use in ELISPOT/ICS.

CPLSKILL Peptide Overview

L-Cysteinyl-L-prolyl-L-leucyl-L-seryl-L-lysyl-L-isoleucyl-L-leucyl-L-leucine (CAS 649760-39-4, molecular formula C₄₁H₇₅N₉O₁₀S, molecular weight 886.2 g/mol) is a synthetic linear octapeptide corresponding to residues 345–352 (sequence CPLSKILL) of the Epstein‑Barr virus (EBV) latent membrane protein 2 (LMP2) [1]. It functions as an HLA‑B*08‑restricted CD8⁺ T‑cell epitope and is primarily employed in T‑cell immunity assays, including IFN‑γ ELISPOT, intracellular cytokine staining (ICS), cytotoxicity, and proliferation assays, for both research and immunotherapy development [1][2]. The peptide is catalogued in the Immune Epitope Database (IEDB) under epitope ID 6808 and is commercially supplied in freeze‑dried form with trifluoroacetate (TFA) counterion at purities ≥90% (HPLC/MS) [3].

HLA-B*08-restricted CD8⁺ T-cell epitope probe

Compatible with IFN‑γ ELISPOT, ICS, cytotoxicity, and proliferation assays

Experimentally validated functional epitope; overlapping peptides inactive

Freeze‑dried research format with TFA counterion

• [1] Tsang ML, Münz C. Cytolytic T lymphocytes from HLA-B8+ donors frequently recognize the Hodgkin's lymphoma associated latent membrane protein 2 of Epstein Barr virus. Herpesviridae. 2011;2:4. doi:10.1186/2042-4280-2-4 View Source
• [2] Immune Epitope Database (IEDB). Epitope Summary: CPLSKILL (Epitope ID 6808). Accessed 2026. View Source
• [3] JPT Peptide Technologies. EBV Latent membrane protein 2 345-352, CPLSKILL, 1 mg Epitope Peptide. Product page. Accessed 2026. View Source

Functional Specificity of CPLSKILL vs. Overlapping Peptides

The LMP2 region spanning residues 340–360 contains several predicted HLA‑B8‑binding octamers and nonamers, including the overlapping peptides LSKILLARL (347–355), ILLARLFL (350–357), and LARLFLYAL (352–360) [1]. Computational prediction tools such as SYFPEITHI and BIMAS assign binding scores to all of these peptides; however, functional validation demonstrates that only CPLSKILL (345–352) reproducibly stimulates LMP2‑specific CD8⁺ T‑cell responses in a large proportion of HLA‑B8⁺ donors [1]. In contrast, the overlapping peptide LMP2₃₄₇₋₃₅₅ (LSKILLARL) is explicitly employed as a negative control in the same study and fails to elicit IFN‑γ secretion or CD69 up‑regulation from CPLSKILL‑specific T‑cell clones [1]. Therefore, procurement based solely on sequence overlap, in silico prediction, or HLA‑binding motif similarity cannot substitute for the experimentally validated, sequence‑specific functional activity of CPLSKILL in HLA‑B8‑restricted T‑cell assays.

Validated Epitope

Overlapping Peptide Substitutes

CPLSKILL (345–352) stimulates IFN‑γ and CD69 up‑regulation

LSKILLARL (347–355) yields no detectable functional response

Majority of HLA‑B8⁺ donors respond in ELISPOT assays

ILLARLFL, LARLFLYAL fail all functional validation

Sequence‑specific T‑cell clone reactivity confirmed

In silico binding predictions do not translate to functional activity

• [1] Tsang ML, Münz C. Cytolytic T lymphocytes from HLA-B8+ donors frequently recognize the Hodgkin's lymphoma associated latent membrane protein 2 of Epstein Barr virus. Herpesviridae. 2011;2:4. doi:10.1186/2042-4280-2-4 View Source

Quantitative Comparison: CPLSKILL vs. Control Epitopes

T-Cell Stimulation: CPLSKILL vs. Predicted LMP2 Peptides

Six LMP2‑derived peptides (172–179, 234–242, 345–352, 347–355, 350–357, 352–360) were predicted by SYFPEITHI and BIMAS to bind HLA‑B8 and were synthesized for functional testing [1]. After one week of co‑culture with peptide‑pulsed autologous dendritic cells, only CPLSKILL (345–352) stimulated LMP2‑specific CD8⁺ T cells in a large proportion of donors, with 8 out of 11 HLA‑B8⁺ donors (73%) meeting the criterion of ≥10 spots over background and a ≥1.5 ratio of vvLMP2 to vvTK‑directed spots [1]. The other five predicted peptides, despite possessing in silico binding scores including SYFPEITHI rank/score 39/20 for LSKILLARL (347–355) and 28/25 for ILLARLFL (350–357), failed to stimulate LMP2‑specific responses [1]. This functional exclusivity establishes CPLSKILL as the only experimentally validated HLA‑B8‑restricted epitope within this region.

T‑cell stimulation

Head‑to‑head

73% donor responsevs0% for 5 predicted peptides

Only CPLSKILL drives functional LMP2 T‑cell response

IFN‑γ ELISPOT; n=11 HLA‑B8⁺ donors

Evidence Dimension	Proportion of HLA-B8+ donors showing LMP2-specific IFN-γ ELISPOT response after DC stimulation
Target Compound Data	8/11 donors (73%) positive; spot-forming cells (SFC) range: 19–581 SFC/5×10⁴ cells
Comparator Or Baseline	LSKILLARL (347–355), ILLARLFL (350–357), LARLFLYAL (352–360), AAAQRKLL (172–179), AYRRRWRRL (234–242): 0/11 donors positive
Quantified Difference	73% vs. 0% donor response rate; only CPLSKILL shows detectable functional T-cell activation
Conditions	IFN-γ ELISPOT assay; HLA-B8+ CD8+ T cells stimulated for 7 days with autologous DCs pulsed with individual peptides, then tested against DCs infected with vvLMP2, vvEBNA3A, or vvTK- (control) recombinant vaccinia viruses

Why This Matters

Procurement of CPLSKILL is mandatory for any experiment requiring functional HLA‑B8‑restricted LMP2 T‑cell stimulation, because overlapping peptides with higher in silico scores do not yield detectable T‑cell activation.

• [1] Tsang ML, Münz C. Cytolytic T lymphocytes from HLA-B8+ donors frequently recognize the Hodgkin's lymphoma associated latent membrane protein 2 of Epstein Barr virus. Herpesviridae. 2011;2:4. Table 1 and Table 2. View Source

Specificity Confirmation: CPLSKILL vs. LSKILLARL Control

Five CPLSKILL‑specific CD8⁺ T‑cell clones were generated from a CD8⁺ T‑cell line with >1% frequency for this specificity [1]. Upon stimulation with 20 μM CPLSKILL peptide plus co‑stimulatory antibodies (CD49d and CD28), all five clones uniformly up‑regulated CD69 (100%), and 30–50% produced IFN‑γ as measured by intracellular cytokine staining and flow cytometry [1]. In contrast, stimulation with medium alone or with 20 μM of the overlapping control peptide LMP2₃₄₇₋₃₅₅ (LSKILLARL) elicited no CD69 up‑regulation and no IFN‑γ production [1]. The same LMP2₃₄₇₋₃₅₅ peptide was also used as a negative control in ELISPOT assays with CPLSKILL‑specific clones against HLA‑B8⁺ lymphoblastoid cell lines (LCL), where it generated no IFN‑γ secretion, confirming that the T‑cell response is strictly sequence‑dependent [1].

Clone specificity

Head‑to‑head

100% CD69⁺, 30–50% IFN‑γ⁺vs0% for LSKILLARL

Absolute sequence‑specific reactivity confirmed

Flow cytometry; 5 T‑cell clones; 20 µM peptide

Evidence Dimension	CD69 up‑regulation and IFN‑γ production in CPLSKILL‑specific CD8⁺ T‑cell clones
Target Compound Data	CPLSKILL (20 μM): 100% CD69 up‑regulation; 30–50% IFN‑γ⁺ cells
Comparator Or Baseline	LMP2₃₄₇₋₃₅₅ (LSKILLARL, 20 μM): 0% CD69 up‑regulation; 0% IFN‑γ⁺ cells; Medium only: 0% CD69; 0% IFN‑γ
Quantified Difference	100% vs. 0% CD69 up‑regulation; 30–50% vs. 0% IFN‑γ positivity
Conditions	Flow cytometry after intracellular IFN‑γ and CD69 staining; clonal CD8⁺ T cells stimulated with 20 μM peptide plus anti‑CD49d and anti‑CD28 co‑stimulatory antibodies

Why This Matters

The exclusive responsiveness of CPLSKILL‑specific T‑cell clones to the 345–352 peptide but not to the overlapping 347–355 peptide demonstrates absolute sequence specificity; substitution with any overlapping peptide would yield false‑negative results in T‑cell assays.

• [1] Tsang ML, Münz C. Cytolytic T lymphocytes from HLA-B8+ donors frequently recognize the Hodgkin's lymphoma associated latent membrane protein 2 of Epstein Barr virus. Herpesviridae. 2011;2:4. Figures 4A and 5. View Source

Ex Vivo CPLSKILL-Specific T-Cell Frequency

Direct ex vivo assessment of whole blood from HLA‑B8⁺ healthy EBV carriers revealed that CPLSKILL‑specific CD8⁺ T‑cell responses were detectable in 4 out of 6 donors (67%), with frequencies ranging from 0.1% to 0.68% of peripheral CD8⁺ T cells [1]. The highest detected response was 0.55% CD69⁺IFN‑γ⁺ plus 0.13% CD69⁺IFN‑γ⁻ cells [1]. In comparison, EBNA3A₃₂₅₋₃₃₃‑specific responses (the immunodominant HLA‑B8‑restricted EBV epitope) were of higher frequency in 5/6 donors, and the frequencies of EBNA3A and LMP2 responses tended to be inversely correlated [1]. This demonstrates that while CPLSKILL responses are subdominant relative to EBNA3A, they are nonetheless reproducibly present in the majority of HLA‑B8⁺ carriers.

Ex vivo frequency

Reported

0.1–0.68% of CD8⁺ T cells

Subdominant epitope detectable ex vivo

4/6 HLA‑B8⁺ healthy carriers; whole blood

Evidence Dimension	Ex vivo frequency of epitope‑specific CD8⁺ T cells in peripheral blood of HLA‑B8⁺ healthy EBV carriers
Target Compound Data	CPLSKILL (LMP2₃₄₅₋₃₅₂): detectable in 4/6 donors (67%); frequency range 0.1–0.68% of CD8⁺ T cells; highest 0.55% CD69⁺IFN‑γ⁺ + 0.13% CD69⁺IFN‑γ⁻
Comparator Or Baseline	EBNA3A₃₂₅₋₃₃₃ (immunodominant HLA‑B8‑restricted epitope): detectable in 5/6 donors (83%); higher frequency in 5/6 donors; inversely correlated with CPLSKILL responses
Quantified Difference	CPLSKILL frequency up to 0.68% vs. EBNA3A higher frequency (inverse correlation); 67% vs. 83% donor detection rate
Conditions	Whole blood stimulation with peptide followed by intracellular IFN‑γ and CD69 staining; flow cytometry; 6 HLA‑B8⁺ healthy EBV carriers

Why This Matters

The quantifiable ex vivo detection of CPLSKILL‑specific T cells in the majority of HLA‑B8⁺ carriers supports its use as a relevant subdominant epitope for immune monitoring studies, where it complements immunodominant epitopes such as EBNA3A to provide a more complete picture of EBV‑specific T‑cell immunity.

• [1] Tsang ML, Münz C. Cytolytic T lymphocytes from HLA-B8+ donors frequently recognize the Hodgkin's lymphoma associated latent membrane protein 2 of Epstein Barr virus. Herpesviridae. 2011;2:4. Figure 2. View Source

CPLSKILL T-Cell Clone Sensitivity Threshold

CPLSKILL‑specific CD8⁺ T‑cell clones were tested for their ability to recognize cognate peptide across a range of concentrations presented by HLA‑B8⁺ lymphoblastoid cell lines (LCL) [1]. The clones recognized peptide concentrations down to 10 nM, as measured by IFN‑γ ELISPOT assays [1]. This threshold is comparable to, though the study does not directly benchmark it against, the sensitivity of immunodominant EBNA3A₃₂₅₋₃₃₃‑specific clones, which also display high‑affinity recognition. The low nanomolar sensitivity indicates that CPLSKILL forms stable pMHC complexes capable of triggering TCR signaling at physiologically relevant antigen densities.

Peptide sensitivity

Class‑level

10 nM recognition threshold

Low nanomolar sensitivity supports pMHC detection

IFN‑γ ELISPOT; HLA‑B8⁺ LCL

Evidence Dimension	Peptide concentration threshold for T‑cell clone recognition by IFN‑γ ELISPOT
Target Compound Data	CPLSKILL (LMP2₃₄₅₋₃₅₂): recognition threshold 10 nM
Comparator Or Baseline	Class‑level: typical HLA‑B8‑restricted immunodominant epitopes (e.g., EBNA3A₃₂₅₋₃₃₃) display similar low nanomolar recognition thresholds; quantitative comparator data not available in this study
Quantified Difference	Absolute threshold: 10 nM (no direct comparator value available; class‑level inference)
Conditions	IFN‑γ ELISPOT; CPLSKILL‑specific CD8⁺ T‑cell clones tested against HLA‑B8⁺ LCL pulsed with titrated peptide concentrations; HLA‑B8⁻ LCL and LMP2₃₄₇₋₃₅₅ peptide as negative controls

Why This Matters

Low nanomolar sensitivity supports the use of CPLSKILL in assays requiring detection of low‑abundance pMHC complexes, such as monitoring of endogenously processed antigen in tumor cells, and indicates that the peptide is suitable for high‑sensitivity immune monitoring applications.

• [1] Tsang ML, Münz C. Cytolytic T lymphocytes from HLA-B8+ donors frequently recognize the Hodgkin's lymphoma associated latent membrane protein 2 of Epstein Barr virus. Herpesviridae. 2011;2:4. Figure 6. View Source

Cytotoxicity of CPLSKILL-Specific T Cells

CPLSKILL‑specific CD8⁺ T cells expanded from HLA‑B8⁺ donors demonstrated dual functionality: they secreted IFN‑γ and mediated cytotoxicity against both peptide‑pulsed target cells and HLA‑B8‑matched LMP2‑expressing Hodgkin's lymphoma cells [1]. In the IEDB, all 2 curated cytotoxicity assays for this epitope returned positive results [2]. The CPLSKILL‑specific clones also recognized two different HLA‑B8⁺ LCL lines (BM‑LCL and SO‑LCL) but not three different HLA‑B8⁻ LCL lines (CM‑LCL, DC‑LCL, LG2), confirming HLA‑B8‑restricted recognition of endogenously processed LMP2 [1]. Although direct quantitative cytotoxicity values (% lysis) are not reported against a comparator peptide in the same study, the class‑level inference is that this peptide is capable of driving tumor‑cell killing, a property not shared by the overlapping non‑functional peptides.

Cytolytic activity

Class‑level

2/2 IEDB cytotoxicity assays positive

Supports tumor‑cell lysis endpoint research

HLA‑B8⁺ LCL and LMP2⁺ Hodgkin's lymphoma cells

Evidence Dimension	Cytolytic activity against HLA‑B8⁺ LMP2‑expressing target cells
Target Compound Data	CPLSKILL‑specific CD8⁺ T cells: kill HLA‑B8⁺ LCL (BM‑LCL and SO‑LCL) and LMP2‑expressing Hodgkin's lymphoma cells; 2/2 positive IEDB cytotoxicity assays
Comparator Or Baseline	LMP2₃₄₇₋₃₅₅ (LSKILLARL): does not stimulate cytotoxicity or IFN‑γ from CPLSKILL‑specific clones; class‑level: other predicted LMP2 peptides (347–355, 350–357, 352–360) failed functional validation
Quantified Difference	Qualitative: CPLSKILL drives tumor‑cell killing (IEDB: 2/2 assays positive); overlapping peptides show no functional activity (IEDB: no positive cytotoxicity assays); exact % specific lysis comparator data not available from this study
Conditions	IFN‑γ ELISPOT and cytotoxicity assays; CPLSKILL‑specific CD8⁺ T‑cell clones tested against HLA‑B8⁺ and HLA‑B8⁻ LCL lines; LMP2‑expressing Hodgkin's lymphoma cells

Why This Matters

The validated cytolytic capacity against HLA‑matched tumor cells qualifies CPLSKILL as a relevant epitope for adoptive T‑cell therapy research and immunotherapeutic targeting of EBV‑associated malignancies such as Hodgkin's lymphoma and nasopharyngeal carcinoma, whereas overlapping peptides lacking this functional validation cannot serve this purpose.

• [1] Tsang ML, Münz C. Cytolytic T lymphocytes from HLA-B8+ donors frequently recognize the Hodgkin's lymphoma associated latent membrane protein 2 of Epstein Barr virus. Herpesviridae. 2011;2:4. Figures 5 and 6. View Source
• [2] Immune Epitope Database (IEDB). Epitope Summary: CPLSKILL (Epitope ID 6808). T Cell Assay data: IFN‑γ release 5/5 positive; cytotoxicity 2/2 positive. View Source

IEDB Curated Evidence for CPLSKILL T-Cell Assays

According to the Immune Epitope Database (IEDB), epitope CPLSKILL (IEDB ID 6808) is supported by 7 curated T‑cell assays across 2 publications, with a 100% positive rate: 5 out of 5 IFN‑γ release assays and 2 out of 2 cytotoxicity assays returned positive results [1]. In contrast, the overlapping peptides LSKILLARL (347–355), ILLARLFL (350–357), and LARLFLYAL (352–360) are not listed in IEDB as independently validated functional epitopes with positive T‑cell assay data. This curated record provides independent, database‑level confirmation of the functional activity and reproducibility of CPLSKILL across independent experimental settings.

IEDB curation

Reported

7/7 T‑cell assays positive

Independent database validation of functional activity

5 IFN‑γ, 2 cytotoxicity; across 2 publications

Evidence Dimension	Curated T‑cell assay positivity rate in IEDB
Target Compound Data	CPLSKILL (IEDB 6808): 5/5 IFN‑γ release positive; 2/2 cytotoxicity positive (7/7 total assays positive)
Comparator Or Baseline	Overlapping LMP2 peptides LSKILLARL, ILLARLFL, LARLFLYAL: no curated positive T‑cell assay data in IEDB (not present as independently validated functional epitopes)
Quantified Difference	100% vs. 0% curated positive assay rate for functional T‑cell activity
Conditions	IEDB curated data; assays include IFN‑γ release (ELISPOT and ICS) and cytotoxicity; human CD8⁺ T cells from HLA‑B8⁺ donors

Why This Matters

The 100% curated positive rate in IEDB across multiple assay types and publications provides independent, third‑party validation of the functional consistency of CPLSKILL, reducing experimental risk for investigators who require a reliably stimulatory HLA‑B8‑restricted EBV LMP2 peptide.

• [1] Immune Epitope Database (IEDB). Epitope Summary: CPLSKILL (Epitope ID 6808). Accessed 2026. View Source

CPLSKILL Application Scenarios

EBV-Specific Immune Monitoring with CPLSKILL

CPLSKILL is the only experimentally validated HLA‑B8‑restricted LMP2 T‑cell epitope within residues 340–360 of EBV LMP2, with 73% (8/11) of HLA‑B8⁺ donors showing detectable responses after DC stimulation and 67% (4/6) showing detectable ex vivo responses [1]. It is used as a peptide stimulus in IFN‑γ ELISPOT assays (typical readout: 19–581 SFC/5×10⁴ cells after in vitro expansion) and in intracellular cytokine staining for CD69 and IFN‑γ detection (30–50% IFN‑γ positivity in antigen‑specific clones at 20 μM peptide) [1]. The commercially available peptide (≥90% purity, freeze‑dried, TFA counterion) is formulated for direct use in these assays [2].

CPLSKILL and LSKILLARL Specificity Control Pair

The CPLSKILL peptide (345–352) and the overlapping LSKILLARL peptide (347–355) form a validated specificity control pair: CPLSKILL stimulates CD69 up‑regulation (100% of clonal cells) and IFN‑γ production (30–50%), while LSKILLARL elicits 0% CD69 and 0% IFN‑γ from CPLSKILL‑specific T‑cell clones at the same concentration (20 μM) [1]. Investigators can use this pair to confirm sequence‑specific T‑cell reactivity in their assays, ruling out non‑specific peptide effects. Both peptides are available from commercial suppliers [2].

CPLSKILL for EBV-Associated Tumor Immunotherapy

CPLSKILL‑specific CD8⁺ T‑cell clones recognize and kill both peptide‑pulsed target cells and HLA‑B8‑matched LMP2‑expressing Hodgkin's lymphoma cells [1]. Given that LMP2 is expressed in EBV‑associated malignancies including Hodgkin's lymphoma and nasopharyngeal carcinoma but not in most healthy tissues, CPLSKILL can be used to expand tumor‑reactive CTLs for adoptive immunotherapy research. The peptide's recognition threshold of 10 nM [1] indicates that endogenously processed LMP2 at low antigen density can be targeted. HLA‑B8 is the third most frequent HLA‑B allele in the United States (8.5% of the population) [1], providing a substantial target patient population for translational studies.

CPLSKILL in Epitope Mapping and Immunodominance

CPLSKILL serves as a subdominant HLA‑B8‑restricted epitope in EBV LMP2, with its ex vivo frequency inversely correlated with the immunodominant EBNA3A₃₂₅₋₃₃₃‑specific response [1]. Investigators examining the hierarchy of EBV‑specific CD8⁺ T‑cell responses can use CPLSKILL in conjunction with EBNA3A₃₂₅₋₃₃₃ and other known HLA‑B8‑restricted epitopes to dissect immunodominance patterns, the impact of antigen expression levels, and the relationship between latent vs. lytic antigen targeting. Unlike the five other predicted LMP2 peptides that failed functional validation, CPLSKILL provides a reliable tool for probing the LMP2‑specific component of the EBV‑specific T‑cell repertoire [1].

Application

Selection Property

Validation Focus

EBV‑specific immune monitoring

HLA‑B8‑restricted validated epitope

Donor response rate and assay reproducibility

Sequence‑specificity control in T‑cell assays

Defined CPLSKILL/LSKILLARL control pair

Peptide‑specific CD69 and IFN‑γ response confirmation

LMP2⁺ tumor cell targeting research

HLA‑B8‑restricted tumor recognition

Cytotoxicity against LMP2‑expressing tumor cells

EBV immunodominance hierarchy studies

Subdominant LMP2 epitope

Inverse correlation with EBNA3A response

• [1] Tsang ML, Münz C. Cytolytic T lymphocytes from HLA-B8+ donors frequently recognize the Hodgkin's lymphoma associated latent membrane protein 2 of Epstein Barr virus. Herpesviridae. 2011;2:4. doi:10.1186/2042-4280-2-4 View Source
• [2] JPT Peptide Technologies. EBV Latent membrane protein 2 345-352, CPLSKILL, 1 mg Epitope Peptide. Product page. Accessed 2026. View Source