Biological assay of peptidoglycans

Abstract

The present invention relates to a biological method for assaying peptidoglycans (PGN) in a sample, particularly a sample of glucose polymers. The PGN assay includes: a) treating the glucose polymer sample by sonication, heating, and/or alkalizing; b) placing the treated sample or a dilution thereof in contact with a recombinant cell expressing an exogenous TLR2 (toll-like receptor 2) and a reporter gene directly dependent on the signaling pathway associated with the TLR2. The reporter gene codes for a colored or fluorescent protein or for a protein the activity of which is measurable with or without a substrate; c) measuring the reporter gene signal; and d) determining the amount of PGN in the sample using a standard curve of the correlation between the amount or PGN and the strength of the reporter gene signal.

Claims

1. A method of assaying a level of peptidoglycan (PGN) contaminants in a sample of glucose polymer manufactured for therapeutic purposes, comprising: obtaining a sample of the glucose polymer manufactured for therapeutic purposes; treating the glucose polymer sample by sonication, heating, and/or alkalization to fragment and disintegrate the PGN contaminants; contacting the treated sample or a dilution thereof with a reporter gene expressing cell line, wherein the reporter gene expressing cell line is a recombinant cell line encoding an exogenous TLR2 receptor (Toll-like Receptor 2) and a reporter gene dependent on a signaling pathway associated with the TLR2 receptor, the reporter gene coding for a colored or fluorescent protein or for a protein whose activity can be measured with or without a substrate; establishing a calibration curve with a PGN standard and the reporter gene expressing cell line; and measuring the reporter gene signal relative to the calibration curve, thereby quantitating a level of contamination in the glucose polymer sample.

2. The method of claim 1, wherein the glucose polymer sample has a concentration of glucose polymers ranging from 5 mg/ml to 50 mg/ml.

3. The method of claim 1, wherein the glucose polymer sample is a solution of maltodextrins.

4. The method of claim 1, wherein said reporter gene is a secreted alkaline phosphatase.

5. The method of claim 1, wherein the cell is a stably transformed HEK-293 cell line expressing a human TLR2.

6. The method of claim 1, wherein the calibration curve is established with a PGN standard prepared with PGNs derived from a bacterium selected from Staphylococcus aureus, Escherichia coli, Micrococcus luteus, Bacillus subtilis and Alicyclobacillus acidocaldarius.

7. The method of claim 6, wherein the PGN standard is a purified and partially digested PGN.

8. The method of claim 1, wherein the sample is diluted to generate a signal of the reporter gene corresponding to a linear portion of the calibration curve.

9. The method of claim 1, wherein the glucose polymer sample is a solution of icodextrin.

10. The method of claim 1 wherein the PGN standard of the calibration curve is a lipopeptide agonist of TLR2.

11. The method of claim 10, wherein the lipopeptide is a PAM.sub.3Cys-Ser-(Lys).sub.4 trihydrochloride.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1: Theoretical curve of the cellular response as a function of increasing concentrations of PGN.

(2) FIG. 2: Calibration curve of the cellular response as a function of the PGN level of S. aureus obtained with the HEK-Blue-hTLR2 cells.

(3) FIG. 3: Response of the HEK-Blue-hTLR2 cells as a function of increasing concentrations of PGN from different bacterial species.

(4) FIG. 4: Response of the HEK-Blue-hTLR2 cells as a function of increasing concentrations of PGN of S. aureus obtained from different batches.

(5) FIG. 5: Structure of PAM.sub.3Cys-Ser-(Lys)4 trihydrochloride (PAM3(cys)).

(6) FIG. 6: Comparison of the responses induced by the PGNs of S. aureus and PAM3(cys) in the HEK-Blue-hTLR2 cells.

(7) FIG. 7: Response of the HEK-Blue-hTLR2 cells as a function of the corrected PGN concentrations.

(8) FIG. 8: Calibration curve of the response of the HEK-Blue-hTLR2 cells as a function of the corrected active PGN concentrations.

EXAMPLES

(9) The assay is based on the specific recognition of PGNs by a line expressing the TLR2 receptor and on the production of an enzyme activity measurable via activation of the signaling pathway associated with TLR2.

(10) Cellular Material

(11) For the experiments relating to this assay, two lines are used:

(12) HEK-Blue hTLR2 line (HEK-TLR2): specific response for the TLR2 ligands, with strong reactivity for the soluble PGNs.

(13) HEK-Blue Null2 line (HEK-Null): nonspecific response connected with a cytotoxic effect of the sample.

(14) The cells are cultured according to the supplier's recommendations (InvivoGen). At 75% confluence, the cells are resuspended at a density of 0.2810.sup.6 cells/mL. Before stimulation, 180 L of the cellular suspension is distributed in the culture wells (96-well plate), or 50 000 cells/well. The cells are then stimulated for 24 h by adding 20 L of the samples of glucose polymer at 37.5% (weight/volume) (i.e. a final dilution of the samples at 3.75%). After 24 h of stimulation, the cellular response is measured by quantification of the enzyme activity produced.

(15) 1Establishment of the Dose-Response Curve

(16) A dose-response curve was constructed by diluting different amounts of PGN standard of S. aureus (FIG. 2) in a solution of uncontaminated icodextrin prepared at 37.5% (weight/volume) (FIG. 2).

(17) The result is a classical curve of cellular response of the sigmoid type.

(18) part (A) corresponds to the responses obtained with low concentrations of PGN, below those giving effective activation of TLR2. This nonlinear zone therefore corresponds to the limit of detection of the method.

(19) part (B) is the most interesting as a linear response is observed. This zone of effective response makes it possible to determine a direct relation between the cellular response and the PGN level. This is therefore the assay zone.

(20) part (C) corresponds to saturation of the cellular response in the presence of excessive concentrations of PGN. There is in fact saturation of the TLR2 receptors.

(21) The standard curve of response of the HEK-TLR2 cells to the PGN of S. aureus has a zone of linearity for concentrations between 0.07 and 10 ng/mL (i.e. between 2 and 267 ng/g of icodextrin).

(22) 2Establishment of the Calibration Curve for Biological Assay of PGNs with an Internal Standard

(23) The dose-response curves were constructed by diluting the PGNs of different bacterial species in a solution of uncontaminated maltodextrin (referenced P-11.11) prepared at 37.5% (weight/volume). The PGNs assayed are extracted from Staphylococcus aureus (Sigma, Cat No 77140), Micrococcus luteus (Sigma, Cat No 53243), Bacillus subtilis (InvivoGen, # tlrl-pgnb2), and Alicyclobacillus acidocaldarius (personal preparation).

(24) The curves obtained are classical curves of the responses observed in the assays performed with a cellular material (bioassay) (FIG. 3). The absorbance values below 0.2 are evidence of PGN concentrations that are too low to induce a cellular response, whereas values above 2 show a plateau effect connected with saturation of the TLR2 receptors. Consequently, only the zone between these two limit values of absorbance allows correlation of the production of SEAP with the amount of PGN present in the samples.

(25) The responses observed show a large variability in the cellular reactivity associated with each type of PGN. In fact, the concentrations giving a response equal to 50% of the maximum response (EC50) are 20 ng/mL for the PGNs of S. aureus and B. subtilis, 1500 ng/mL for M. luteus, and more than 2000 ng/mL for those extracted from A. acidocaldarius and E. coli K12.

(26) However, these differences were expected, since the PGNs have different structures depending on their bacterial origin, which is responsible for large variations in inflammatory reactivity. These observations emphasize the importance of defining an internal standard so as to be able to express the results in equivalent units of PGN.

(27) Another factor likely to alter the response of the HEK-TLR2 cells is the size of the PGNs, which will influence their solubility and reactivity with respect to TLR2. Thus, the procedure for purification of these macromolecules may have a considerable influence on the response of the cells, since the conditions of extraction could alter the size of the PGNs, or even cause partial degradation. To test this hypothesis, the assays were reproduced with 3 separate batches of PGNs extracted from S. aureus: 2 Sigma batches (Cat No 77140: batch 1, 0001442777; batch 2, BCBH7886V) and 1 InvivoGen batch (# tlrl-pgnsa).

(28) The results show variability of reactivity between the three batches (FIG. 4). In fact, the EC50 values are 4, 20 and 400 ng/mL respectively for the three batches. These data indicate that there is a risk that PGNs extracted from the same bacterial species might show differences in reactivity, even if the batches were obtained from the same supplier and were extracted beforehand by the same procedure. It therefore seems necessary to introduce an internal standard for the calibration curve, so as to avoid errors relating to the variability of the PGNs and to express the results as amount of active PGN.

(29) PAM.sub.3Cys-Ser-(Lys)4 trihydrochloride (PAM3(cys); FIG. 5) is a triacylated synthetic lipopeptide that mimics the structure of the bacterial lipopeptides and acts as a strong agonist of TLR2. Being of homogeneous structure, it is often used as positive control for calibrating the responses of the cells expressing the TLR2 receptor.

(30) The experiments were therefore reproduced replacing PGN with PAM3(cys) in our tests. As expected, the HEK-TLR2 cells show strong reactivity with respect to this compound. Moreover, the shape of the dose-response curve is similar to those obtained in the presence of PGN, with EC50 estimated at 10 ng/mL (FIG. 6). These results indicate that PAM3(cys) induces responses equivalent to those of the most reactive PGNs, but in contrast to the latter, it does not display structural variability. Consequently, this synthetic lipopeptide can be used for calibrating the batches of PGN and for establishing a standardized calibration curve, which will allow the results to be formulated in amounts of active PGN, i.e. in amounts of PGN giving TLR2 responses identical to those obtained with the same amounts of PAM3(cys).

(31) Each batch of PGN is calibrated relative to PAM3(cys) by comparing the slopes of the linear portions of each dose-response curve, and calculating a correction factor for superimposing the curves of the PGNs on that of PAM3(cys). In the example presented in FIG. 6, the correction factors were estimated at 0.4, 2 and 40 for batches 1, 2 and 3, respectively. This means that 2.5 times less PGN from batch 1 is required for obtaining responses identical to those induced by PAM3(cys), but 2 times more PGN from batch 2 and 40 times more PGN from batch 3. After correcting the raw quantities of PGN, it can be seen that all the points are aligned on one and the same curve, which can be superimposed on that obtained with PAM3(cys) (FIG. 7). Consequently, using the internal standard makes it possible to obtain corrected concentrations for all the batches of PGN and establish a dose-response curve calibrated for active PGN.

(32) By applying this method, the standard curve of response of the HEK-TLR2 cells has a zone of linearity for concentrations of active PGN between 0.5 and 200 ng/mL (FIG. 8), or between 13 and 5400 ng/g of glucose polymers.