We have already described how limulus amebocyte lysate (LAL) reacts to both endotoxins and (1→3)-β-D-glucan. When β-glucan is present during an assay, endotoxin sometimes cannot be accurately measured. Consequently, there have been strong calls for LAL reagents that are specific to endotoxins.

One response to these calls was the development by Obayashi et al.¹⁾ of an endotoxin-specific reagent called Endospecy (Seikagaku Corporation). Recognizing the demand for a product that is unlikely to become contaminated during formulation and is compatible with gel-clot and toxinometer methods, we decided to follow the work of Obayashi et al. by designing and marketing a line of endotoxin-specific reagents called the Limulus ES test Wako that is based on a different principle.

Information about the technique conceived by the Obayashi group can be reviewed through the reference given below. This post will outline how we developed our proprietary reagent.

The effects of β-glucans on endotoxin measurements during LAL tests are complicated. As explained in episode of β-glucan, one primary reason is that β-glucan and endotoxin react with LAL through different activation mechanisms. Because the slopes of the calibration curves for endotoxin and β-glucan are different, gaining quantitative data from a mixture of the two substances is not as simple as saying one plus one equals two.

For instance, when measured with our HS-type LAL (which reacts to endotoxin and β-glucan) and ES-type LAL (which reacts only to endotoxin), the difference between the measured values from the two reagents does not necessarily indicate the quantitative level of β-glucan. Although a clearly higher result from the HS-type reagent than ES-type suggests the presence of β-glucan in a sample, quantification may be challenging.

In 1981, Kakinuma et al. published a report on the reaction of LAL with carboxymethylated β-glucan (CMPS) ²⁾. They found that activation of LAL by CMPS occurs at an optimum CMPS concentration, but the reaction is inhibited when CMPS concentrations are higher. They also learned that a large amount of endotoxin (10 ng/mL) can still activate LAL in the presence of a high concentration of CMPS.

Although they mentioned the effects of β-glucans on LAL tests, they did not consider measuring endotoxin in the absence of β-glucan interference.

The study by Kakinuma et al. gave us the idea to quantify endotoxin in a situation where an excess amount of β-glucan completely suppresses its reaction with LAL. In order to realize this concept, it was necessary to expand on the experiment by Kakinuma et al. and answer the following questions:

1. Can β-glucan activation of LAL be completely suppressed by adding more β-glucan?
2. To what extent does a large excess of β-glucan impact endotoxin quantification?

If excess β-glucan fails to entirely prevent LAL activation, then endotoxin quantification and reagent stability cannot be guaranteed. Moreover, if the addition of more β-glucan interferes with endotoxin measurements so that endotoxin is not detected accurately, then this technique would be impractical. Fortunately, the results of our study confirmed the feasibility of formulating an endotoxin-specific reagent ³⁾.

There was one more issue that had to be resolved before commercialization: an endotoxin-free preparation of β-glucan had to be developed. This post will refrain from giving any details concerning this issue because some manufacturing processes are confidential. Suffice it to say, the endotoxin-free status was achieved by modifying β-glucan under basic conditions to produce a water-soluble derivative.

While unrelated to endotoxin contamination, the fact that a water-soluble β-glucan can be used has great implications for the formulation of reagents required for the principle described above.

In summary, it was found that an endotoxin-specific reagent can be prepared by simply adding β-glucan. This technique is advantageous not only because it is compatible with gel-clot, kinetic turbidimetric, and chromogenic assays, but also because it is both easy to implement and less prone to contamination.

References

1. Obayashi, T. et al. : Clin. Chim. Acta., 148, 55 (1985).
2. Kakinuma, A. et al. : Biochem. Biophys. Res. Commun., 101, 434 (1981).
3. Tsuchiya, T. (1990). Japanese Journal of Bacteriology, 45, 903 (1990).