Application of a Recombinant Three-Factor Chromogenic Reagent, PyroSmart, for Bacterial Endotoxins Test Filed in the Pharmacopeias

Abstract

Assays using lysate reagents prepared from horseshoe crab hemocyte extract (limulus amoebocyte lysate, LAL) are commonly and widely used to detect and measure endotoxin in parenteral drugs and medical devices. However, lysate reagents suffer from lot-to-lot variations leading to possible fluctuations in testing. Also, this continued usage of lysate reagents leads to the possible decline of the horseshoe crab population. Recently, a new recombinant chromogenic reagent, PyroSmart, consisting of three recombinant factors was introduced to the market. There are now three recombinant products; two with recombinant factor C reagents and PyroSmart with the complete recombinant LAL system. We evaluated the applicability of the reagent to the harmonized bacterial endotoxins test in the United States, European and Japanese pharmacopeias. The recombinant product showed equivalent potency of thirteen endotoxins from different bacterial strains to conventional chromogenic lysate reagents as long as their assay modes are identical. All analytical characteristics or assay parameters of the reagent satisfied the acceptance criteria which are set for the use for the bacterial endotoxins test filed in the pharmacopeias. All of 109 parenteral drugs tested can be measured with PyroSmart within respective maximum allowable dilutions. The lot-to-lot variation in recovery of endotoxin added in the parenteral drugs for PyroSmart was equal to or less than those of six limulus lysate reagents. In conclusion, the present study suggests that the recombinant reagent, PyroSmart, provide a good alternative to the LAL reagents with better lot-to-lot variation.

INTRODUCTION

Parenteral drugs are strictly managed by pharmacopeias in all regions, including the United States, the European Union, and Japan, to ensure their quality and to avoid any health hazards. In particular, the level of pyrogen must be controlled as it potentially causes life-threating septic shock.^1–5) Bacterial endotoxin, (also called lipopolysaccharide; LPS), of Gram-negative bacteria is the pyrogen of main concern to the parenteral drugs in pharmaceutical factories as it is a common contaminant and shows powerful pyrogenicity.^1,2)

Bacterial endotoxins are detected and quantified by lysate reagents, which are prepared from amoebocyte extracts (limulus amoebocyte lysate; LAL) from a natural resource; horseshoe crab.^6–8) Lysate reagents are composed of three protease zymogens, factor C, factor B, and the proclotting enzyme, as well as a coagulogen to form the cascade reaction.⁹⁾ The test using limulus lysate reagents is called Bacterial Endotoxins Test (BET), and is filed in the pharmacopeias.^10–12)

The reliance of the lysate reagents on a natural resource implies three issues. The first is the potential decrease in the population of the horseshoe crab due to harvesting by commercial fishing to bait traps for eels and conch.^13–15) The second is the lot-to-lot variation of limulus lysate reagents possibly due to geographical and seasonal differences.^16–18) The third is a nonspecific reaction derived from an alternative factor G pathway in LAL, reacting with (1 → 3)-β-D-glucan.^8,9) This factor G pathway may lead to a false-positive test result. To address these issues, some products have been marketed, including PyroGene and EndoZyme, which use recombinant factor C alone and a fluorogenic substrate.^19,20) PyroSmart is characterized by three recombinant factors including factor C, factor B and the proclotting enzyme and a chromogenic substrate.²¹⁾

Several reports regarding evaluations of PyroGene have been published.^20,22,23) In these evaluations the equivalency of the endotoxin levels in some parenteral drugs by PyroGene and the current limulus lysate reagents was discussed. However, equivalency between all currently available recombinant reagent products and the existing limulus lysate reagents in measuring endotoxin levels in the many parenteral drugs listed in pharmacopeias has not been fully investigated.

In this report, the potencies of various rough and smooth forms of endotoxins from different bacterial strains using PyroSmart were compared against six lysate reagents. In addition, we assessed the performance of PyroSmart’s analytical characteristics. Based on the results obtained, the applicability of the PyroSmart to BET is shown. The equivalency of measuring endotoxin with PyroSmart and current limulus lysate reagents is discussed.

MATERIALS AND METHODS

Materials

Reference standard endotoxin of the U.S. Pharmacopeia (USPRSE) and the Japanese Pharmacopoeia (JPRSE) was purchased from the U.S. Pharmaceutical Convention (MD, U.S.A.) and the Pharmaceutical and Medical Device Regulatory Science Society of Japan (PMRJ; Osaka, Japan), respectively. Endotoxins from Escherichia coli O55:B5, E. coli O111:B4, Salmonella minnesota R595 Re, S. typhimurium were purchased from List Biological Laboratory, Inc. (CA, U.S.A.). Endotoxins from E. coli O127:B8 and Shigella flexneri were from BD Difco (MI, U.S.A.). Endotoxins from E. coli O128:B12, E. coli J5, E. coli F583 Rd2, S. typhosa and Pseudomonas aeruginosa 10 (purified by gel filtration) were from Sigma-Aldrich (MO, U.S.A.). Pachyman, (1 → 3)-β-D-glucan (BG), was prepared from Poria cocos.¹³⁾ Water for injection purchased from Otsuka Pharmaceutical Factory, Inc. (Tokushima, Japan) was used as water for the BET. Kinetic-QCL and PyroGene were purchased from Lonza (MD, U.S.A.), EndoZyme and EndoZyme II were purchased from bioMérieux (Munich, Germany), Limulus ES-II and Limulus Color KY Test were obtained from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan), Endochrome-K was purchased from Charles River Laboratories (MA, U.S.A.) and Pyrochrome with Glucashield buffer, Endospecy ES-50M and PyroSmart were obtained from Seikagaku Corporation (Tokyo, Japan).

Endotoxin Assay

Measurements of endotoxins with the lysate reagents, Kinetic-QCL, Limulus ES-II, Limulus Color KY Test, Endochrome-K, Endospecy ES-50M and Pyrochrome reconstituted with Glucashield buffer, and also the recombinant reagents, PyroGene, EndoZyme and PyroSmart, were performed according to their Instruction for Use (IFU). Limulus ES-II is turbidimetric, PyroGene and EndoZyme are fluorogenic, and all the other reagents are chromogenic reagents. Both rate and onset time assays are the method to measure the rate of color development of the reaction mixture and to measure the time (onset time) needed to reach a predetermined absorbance, respectively. To calculate the kinetics, an onset time assay was used for Kinetic-QCL, Limulus ES-II, Limulus Color KY Test, Endochrome-K and Pyrochrome, and rate assay for Endospecy ES-50M, and endpoint assays were employed for PyroGene and EndoZyme, and both rate and onset time assays were recorded for PyroSmart.

Potencies of Various Endotoxins with Lysate Reagents and Recombinant Reagents

USPRSE was used as reference standard and was diluted to prepare a series of standard solutions of 0.1, 0.05, 0.025, 0.0125, and 0.00625 endotoxin unit (EU)/mL and a series of standard solutions of 50, 5, 0.5, 0.05 and 0.005 EU/mL for rate and onset time assay mode, respectively. Endotoxins derived from Escherichia coli O55:B5, E.coli O111:B4, E.coli O127:B8, E.coli O128:B12, E.coli J5, Shigella flexneri and Salmonella enterica, were dissolved into water. The endotoxins from E. coli F583 Rd2, S. minnesota R595 Re, S. typhrimrium, S. typhosa and Pseudomonas aeruginosa 10, were dissolved into 0.1%(v/v) triethylamine (TEA). Subsequently, the dissolved endotoxins were diluted with water for BET to obtain a two-fold or ten-fold dilution series for tests in the endotoxin assays with various reagents. The potency, EU/ng, of each dilution of endotoxins was calculated by dividing EU/mL by concentration, ng/mL at each dilution of endotoxins. The mean potencies of endotoxins were calculated by averaging the potencies of endotoxin dilutions within the range of the standard solutions.

Assessments of the Analytical Characteristics of the Reagent

We evaluated PyroSmart with respect to analytical characteristics, accuracy, precision, specificity, quantitation limit, linearity, and range, which are required to be evaluated in quantitative tests for impurities’ content in the ICH Q2 guideline.²⁴⁾ We used both rate and onset time assay modes to assess the analytical characteristics of the recombinant chromogenic reagent. The acceptance criteria described in the Guideline on Bioanalytical Method (Ligand Binding Assay) Validation in Pharmaceutical Development,²⁵⁾ which is used to evaluate measuring methods of analytes in a given biological matrix was referred since the ICH Q2 guideline as well as the Japanese Pharmacopoeia (JP), United States Pharmacopeia (USP), European Pharmacopoeia (EP), and their related guidelines do not describe any acceptance criteria on repeatability, intermediate precision, reproducibility, and quantitation limit. The specificity was assessed based on the reactivity of the reagent with BG, which is well-known as a false positive substance to lysate reagents. We assessed these analytical characteristics except for specificity by changing three factors, analyst, equipment, and reagent lot. In practice, two analysts used two microplate readers and three lots of the reagent to conduct independently a total of 12 measurements.

Linearity—For the rate assay, a series of JPRSE dilutions of 0.1, 0.05, 0.025, 0.0125, and 0.00625 EU/mL were prepared and then measured in six replicates with the reagent. For the onset time assay, a series of USPRSE dilutions of 50, 5, 0.5, 0.05, and 0.005 EU /mL were prepared and then measured in eight replicates with the reagent. According to the regression analysis, the correlation coefficients for the rate and onset time assay modes were determined based on the results of twelve measurements.

Accuracy—Accuracy was assessed by calculating the recoveries of endotoxin based on the average of concentrations of endotoxin that had been measured in six replicates and eight replicates according to the rate and onset time methods, respectively, at each concentration of endotoxin on the regression line in the twelve measurements.

Repeatability—The coefficient of variation (CV) was calculated based on the concentration of endotoxin that was measured at each endotoxin concentration on the regression line in the twelve measurements described in the section of linearity assessment according to the rate and onset time assay, respectively.

Intermediate precision—The standard deviation (SD) was calculated based on twelve measurements in the section of linearity assessment in the rate assay mode, while the SD was calculated from each logarithmically converted value of the endotoxin concentration on the regression line in the twelve measurements described in the section of linearity assessment in the onset time mode. The 90% confidence interval (CI) for CV was calculated according to the following equation: [(n − 1) SD²/χ² (n − 1, 0.05) ≤ σ² ≤ (n − 1) SD²/χ² (n − 1, 0.95)], where “σ” expresses the population standard deviation.

Reproducibility—Reproducibility was assessed in the onset time assay mode using a total of eighteen measured concentrations of endotoxin in two different laboratories consisting of the twelve measurements described in the section for linearity assessment and further six measurements in another institution, where one analyst measured three lots of the reagent with one microplate reader at two different days. Subsequently, SD was calculated from the abovementioned all eighteen measurements, and the 90% CI for the CV was calculated as described above. Furthermore, the eighteen endotoxin concentrations were converted logarithmically, and two-way repeated measures ANOVA was made (variable 1: two institutions, variable 2: five concentrations of endotoxin).

Range—The range is the interval between the upper and lower concentration of endotoxin in the sample for which it has been demonstrated that the analytical procedure has acceptable levels of accuracy, precision (repeatability and intermediate precision), and linearity.²⁴⁾

Quantification limit (QL)—In the rate assay mode QL was calculated based on the slope of the standard curve around QL and on SD of the absorbance change rates in the water for BET as the blank sample: (QL = 10 SD/slope). A series of JPRSE dilutions of 0.01, 0.005, 0.0025 and 0.00125 EU/mL were prepared. The series of JPRSE dilutions and blank sample were measured in five and ten replicates, respectively. Three measurements with three PyroSmart lots—a total of nine measurements—were conducted. The 95% CI for the mean of the QLs was calculated. In the onset time, QL was determined as a minimum concentration where the acceptance criteria were satisfied for both accuracy and precision among endotoxin concentrations described above.

Specificity—To verify interference by (1 → 3)-β-D-glucan (BG), two series of USPRSE dilutions (0.1, 0.05, 0.025 and 0.0125 EU/mL) with and without the addition of pachyman as BG (5 µg/mL)²⁶⁾ were measured in three replicates, and the mean of the endotoxin concentrations were calculated. Regression analysis was conducted on measured endotoxin concentrations with and without BG by using three lots of PyroSmart.

Recovery of JPRSE added into various parenteral drugs was also assessed as one item of specificity. Endotoxin was added to the previously diluted drugs as well as water for the BET setting their endotoxin concentration as the middle point of standard solutions. The recoveries of the endotoxin added to the diluted drugs were calculated from the concentration found in the endotoxin-added drugs after subtracting the endotoxin concentration found in the drugs without addition. Diluted drugs under this test are judged to be free from interfering factors when the calculated recoveries of the endotoxins added to diluted drugs are within 50 and 200%. Minimum dilution factors are expressed as the Non-Interfering-Dilution (NID) where the drugs show neither inhibition nor enhancement in the test. In onset time assay, measurement time and threshold absorbance (onset OD) were optimized to appropriately measure endotoxins in the drugs.

RESULTS

Potency of Endotoxins from Different Bacterial Strains Measured with PyroSmart and Limulus Lysate Reagents

The equivalence of product performance between a recombinant chromogenic reagent, PyroSmart, and limulus lysate reagents was evaluated from potencies of the endotoxins from thirteen strains of Gram-negative bacteria using either the rate or the onset time assay. These data are shown in Fig. 1. All potencies of thirteen endotoxins with PyroSmart fell in the range between 50% of minimum and 200% of maximum of those tested with the three limulus lysate reagents. The endotoxins that were dissolved initially in TEA, specifically from E. coli strain F583 Rd2 and the S. minnesota strain R595 Re which has a short-sugar chain structure (rough form), showed relatively large reagent-to-reagent variability when compared with other endotoxins. On the other hand, the relative potencies of JPRSE with a long-sugar chain structure (smooth form) which had been well purified and dissolved in water for the BET were almost equivalent using PyroSmart and limulus lysate reagents. All of the CVs which were calculated from the potencies of endotoxins with the three lots of PyroSmart were below 10%, except for Pseudomonas aeruginosa 10 (12.7%).

Fig. 1. Comparison of the Reactivity of Different Endotoxins in the Assays with Three Lots of Recombinant Chromogenic Reagent (PyroSmart) and One Lot of Three Lysate Reagents

Reactivity is expressed as EU/ng using USP reference standard endotoxin (RSE) as a reference. Error bars indicate standard deviation in the results of 3 lots.

In addition, the potencies of endotoxins measured with two kinds of reagents were compared in regression analysis. The analysis was done after potencies were transformed logarithmically in nine combinations (combinations A to F, between PyroSmart and lysate reagents; combinations G to I, among lysate reagents; Table 1, Supplementary Fig. 1). In the case of combination A (PyroSmart and Endospecy; both in the rate assay mode) and combination E (PyroSmart and Kinetic-QCL; both in the onset time assay mode) the 95% CI of the slope included “1,” and the 95% CI for the y-intercept included “0,” indicating that the potencies of thirteen endotoxins that were measured with PyroSmart and lysate reagents in combinations A and E are statistically equivalent. In addition, the correlation coefficients in the combinations A and E were 0.936 and 0.947, respectively, showing good correlation in the two combinations.

Table 1. Regression Analyses of the Potency of Endotoxins with a Recombinant Chromogenic Reagent and Lysate Reagents

Combination	X Axis	Y Axis	Slope			Y-Intercept			Correlation coefficient
			Slope	95% CI		Y-Intercept	95% CI
				Lower limit	Upper limit		Lower limit	Upper limit
A	Endospecy ES-50M1),†	PyroSmart1),†	0.968	0.726	1.210	0.142	−0.057	0.340	0.936
B	Kinetic-QCL1),‡	PyroSmart1),†	0.767	0.574	0.961	0.016	−0.214	0.246	0.935
C	Limulus ES-II2),‡	PyroSmart1),†	0.889	0.444	1.333	0.102	−0.304	0.508	0.799
D	Endospecy ES-50M1),†	PyroSmart1),‡	1.463	1.059	1.866	−0.050	−0.382	0.282	0.923
E	Kinetic-QCL1),‡	PyroSmart1),‡	1.191	0.924	1.458	−0.274	−0.591	0.043	0.947
F	Limulus ES-II2),‡	PyroSmart1),‡	1.277	0.527	2.026	−0.054	−0.738	0.630	0.749
G	Endospecy ES-50M1),†	Kinetic-QCL1),‡	1.217	1.000	1.434	0.197	0.018	0.375	0.966
H	Endospecy ES-50M1),†	Limulus ES-II2),‡	0.845	0.588	1.102	0.225	0.013	0.436	0.909
I	Kinetic-QCL1),‡	Limulus ES-II2),‡	0.611	0.337	0.885	0.179	−0.146	0.504	0.828

United States Pharmacopoeia Reference Standard Endotoxin was used as reference. 1) chromogenic reagent; 2) turbidimetric reagent; ^†: rate assay mode; ^‡: onset time assay mode; CI: confidence interval.

Analytical Characteristics of PyroSmart

Table 2 shows the comparison of the acceptance criteria to the results generated by PyroSmart.

Table 2. Assessment of Analytical Characteristics According to ICH Q2 Guideline²⁴⁾

Analytical characteristics		Results				Acceptance criteria
		Rate assay mode		Onset time assay mode
1	Linearity (correlation coefficient: absolute value)	0.00625–0.1 EU/mL		0.005–50 EU/mL		|r| ≥ 0.980
		0.999 (minimum)		1.000
		1.000 (maximum)
2	Accuracy (recovery)	EU/mL	Min–Max (%)	EU/mL	Min–Max (%)
		0.00625	72.7–99.4	0.005	94.7–103.6	50–200%
		0.0125	94.2–101.3	0.05	91.6–103.1
		0.025	100.0–105.6	0.5	99.1–108.2
		0.05	100.3–105.4	5	100.1–109.4
		0.1	98.5–99.8	50	92.4–99.3
3	Precision
	3–1 Repeatability (CV)	EU/mL	Min–Max (%)	EU/mL	Min–Max (%)
		0.00625	3.1–11.0	0.005	5.9–15.7	CV ≤25% at lowest conc.
		0.0125	1.4–7.2	0.05	5.7–14.9	CV ≤20% at the other conc.
		0.025	1.5–6.5	0.5	6.4–11.4
		0.05	1.4–5.4	5	3.3–14.8
		0.1	0.6–4.3	50	3.1–14.2
	3–2 Intermediate precision (90% CI for CV)	EU/mL	Lower limit–Upper limit (%)	EU/mL	Lower limit–Upper limit (%)
		0.00625	7.3–15.2	0.005	0.4–0.7	CV ≤25% at lowest conc.
		0.0125	1.4–2.9	0.05	0.7–1.5	CV ≤20% at the other conc.
		0.025	1.2–2.5	0.5	2.9–6.1
		0.05	1.3–2.6	5	1.2–2.5
		0.1	0.3–0.7	50	0.4–0.8
	3–3 Reproducibility (90% CI for CV)	Not assessed.		EU/mL	Lower limit–Upper limit (%)
				0.005	0.6–1.1	CV ≤25% at lowest conc.
				0.05	1.0–1.8	CV ≤20% at the other conc.
				0.5	4.0–7.1
				5	1.0–1.8
				50	0.5–0.9
4	Range	0.00625–0.1 EU/mL		0.005–50 EU/mL		Precision, accuracy and linearity are to be suitable level.
5	Quantitation limit	95% CI		at 0.005 EU/mL		The lowest concentration of Et can be quantitatively determined with suitable precision and accuracy.
		Lower limit-upper limit		Accuracy: 94.7–103.6%
		0.0008–0.0016 EU/mL		Repeatability: 5.9–15.7%
6	Specificity (reactivity with BG) USPRSE was used.	Results from the regression analysis (95% CI) Intercept: −0.133 to 0.126 Slope: 0.991–1.029		—		Not reactive to BG

Japanese Pharmacopoeia Reference Standard Endotoxin was used for rate assay mode except for specificity, and United States Pharmacopeia Reference Standard Endotoxin (USPRSE) was used for onset time assay mode. EU, endotoxin unit; CV, coefficient of variation; CI, confidence interval; BG, (1→3)-β-D-glucan.

Linearity—Typical standard curves for endotoxin quantitation obtained with PyroSmart in the rate and onset time assays are shown in Fig. 2. Both in the rate and onset time assays, the correlation coefficients calculated from all standard curves satisfied the acceptance criteria (|r| ≥ 0.980) that are described in the preparatory test of the BET.

Fig. 2. Typical Standard Curves for Quantification of Endotoxin Obtained with Recombinant Chromogenic Reagent (PyroSmart)

Absorbance change rate (mAbs/min) obtained according to the rate assay mode (Panel A) or the onset time (second) obtained according to onset time mode (Panel B) are plotted against concentration of JP or USP reference standard endotoxin (RSE). Error bars show standard deviation in six replicates (Panel A) and eight replicates (Panel B) measurements.

Accuracy—In either the rate or onset time assay, the accuracy value, namely recovery, of added endotoxin satisfied the acceptance criteria for the test of interfering factors as described in the BET, i.e., 50–200%.

Precision—Repeatability, Intermediate precision and Reproducibility all satisfied their acceptance criteria as shown in Table 2.

Range—The range was found to be 0.00625–0.1 EU/mL for the rate assay and 0.005–50 EU/mL for the onset time method.

Quantitation limit—In the rate assay mode, the mean of the quantitation limit was 0.0012 EU/mL. The upper limit of the 95% CI for the endotoxin concentrations in the blank sample was 0.0002 EU/mL, while the lower limit of the 95% CI for the JPRSE 0.00125 EU/mL was 0.00126 EU/mL. Therefore, the ranges of endotoxin concentrations of in the blank sample and JPRSE did not overlap. Using onset time, both accuracy and precision at 0.005 EU/mL of endotoxin satisfied the acceptance criteria, indicating the lower limit of quantitation for the onset time method is 0.005 EU/mL.

Reactivity to (1 → 3)-β-D-glucan (BG)—Regression analysis for the results of three lots with and without the addition of BG was conducted (Fig. 3). The 95% CI for the y-intercept and the slope of the regression formula with the three lots contained “0” and “1,” respectively, indicating there is no statistical difference between the endotoxin concentrations with and without BG.

Fig. 3. Reactivity of Recombinant Chromogenic Reagent (PyroSmart) to Endotoxin in the Presence or in the Absence of (1 → 3)-β-D-glucan (BG)

Regression analysis was done as described in Materials and Methods.

Ability to assess endotoxin in the presence of injectable drugs—We examined the interference to the test by examining the recovery of endotoxins added to 109 out of 118 available injectable drugs listed in the monographs of Japanese Pharmacopoeia. In both the rate and onset time assays, all minimum dilution factors expressed as NIDs of the 109 injectable drugs, which were measured with PyroSmart, were smaller than the MVDs of respective drugs (Table 3). NIDs of 109 injectable drugs which were measured with three lysate reagents were also smaller than their MVDs.

Table 3. Non-interfering-Dilution (NID) of Injectable Drugs Tested with the Recombinant Chromogenic Reagent

ID	Parenteral drug	Stock solution conc.	Release limit	NID (upper line)/MVD (lower line)
				PyroSmart		Endospecy	Kinetic-QCL	Limulus ES-II
				Rate	Onset time	Rate	Onset time	Onset time
1	Aciclovir injection	25 mg/mL	0.5 EU/mg	2	8	4	8	8
				2000	2500	2000	2500	1600
2	Ascorbic acid injection	100 mg/mL	0.15 EU/mg	4	2	4	8	4
				2400	3000	2400	3000	1920
3	Aztreonam for injection	333 mg/mL	0.10 EU/mg1)	32	64	16	128	32
				5333	6667	5333	6667	4267
4	Atropine sulfate injection	0.5 mg/mL	75 EU/mg	1	1	1	4	1
				6000	7500	6000	7500	4800
5	Amikacin sulfate injection	100 mg/mL	0.50 EU/mg1)	16	16	2	64	4
				8000	10000	8000	10000	6400
6	Amikacin sulfate for injection	50 mg/mL	0.50 EU/mg1)	16	16	1	64	2
				4000	5000	4000	5000	3200
7	Aminophylline injection	25 mg/mL	0.6 EU/mg	16	16	8	4	8
				2400	3000	2400	3000	1920
8	Amphotericin b for injection	4.2 mg/mL	3.0 EU/mg1)	8	8‡	4	32	64
				2000	2500	2000	2500	1600
9	L-Arginine hydrochloride injection	100 mg/mL	0.50 EU/mg1)	8	8	4	16	4
				80	100	80	100	64
10	Alprostadil injection	0.005 mg/mL	10 EU/mL	8	256	16	1	256
				1600	2000	1600	2000	1280
11	Arbekacin sulfate injection	50 mg/mL	0.50 EU/mg1)	4	4‡	2	32	2
				4000	5000	4000	5000	3200
12	Alendronate sodium injection	2.5 mg/mL	119 EU/mg	2	2	2	16	8
				47600	59500	47600	59500	38080
13	Ampicillin sodium for injection	250 mg/mL	0.075 EU/mg1)	8	32‡	8	16	32
				3000	3750	3000	3750	2400
14	Sodium iotalamate injection	66.8 %	3.4 EU/mL	32	64	16	32	32
				544	680	544	680	435
15	Isepamicin sulfate injection	200 mg/mL	0.50 EU/mg1)	16	32	4	128	8
				16000	20000	16000	20000	12800
16	Isoniazid injection	50 mg/mL	0.50 EU/mg	4	4	4	4	4
				4000	5000	4000	5000	3200
17	Idarubicin hydrochloride†		8.9 EU/mg1)
18	Idarubicin hydrochloride for injection	1 mg/mL	8.9 EU/mg1)	128	512	128	16	8
				1424	1780	1424	1780	1139
19	Imipenem and cilastatin sodium for injection	25 mg/mL	0.25 EU/mg1)	8	16	1	4	16
				1000	1250	1000	1250	800
20	Indigocarmine injection	4 mg/mL	7.5 EU/mg	4	16	4	16	32
				4800	6000	4800	6000	3840
21	Insulin human (genetic recombination)	100 unit2)/mL	0.8 EU/unit2)	4	2	2	8	2
				12800	16000	12800	16000	10240
22	Edrophonium chloride injection	1 mg/mL	15 EU/mg	8	8	4	8	4
				2400	3000	2400	3000	1920
23	Ephedrine hydrochloride injection	40 mg/mL	7.5 EU/mg	8	20	4	8	8
				48000	60000	48000	60000	38400
24	Ergometrine maleate injection	0.2 mg/mL	1500 EU/mg	1	2	1	32	1
				48000	60000	48000	60000	38400
25	Calcium chloride injection	55.5 mg/mL	0.30 EU/mg	32	32	64	32	16
				2664	3330	2664	3330	2131
26	10% Sodium chloride injection	10 %	3.6 EU/mL	2	2	4	16	2
				576	720	576	720	461
27	Oxytocin injection	1 unit/mL	10 EU/unit	2	16	2	8	2
				1600	2000	1600	2000	1280
28	Ozagrel sodium for injection	10 mg/mL	3.7 EU/mg	1	1	2	2	4
				5920	7400	5920	7400	4736
29	Fructose injection	20 %	0.5 EU/mL	4	4	2	4	2
				80	100	80	100	64
30	Xylitol injection	5 %	0.50 EU/mL	1	1	1	1	1
				80	100	80	100	64
31	Sodium citrate injection for transfusion	100 mg/mL	5.6 EU/mL	1	1	4	64	4
				896	1120	896	1120	717
32	Clindamycin phosphate injection	150 mg/mL	0.1 EU/mg1)	128	64	16	32	32
				2400	3000	2400	3000	1920
33	Chlorpheniramine maleate injection	2 mg/mL	8.8 EU/mg	2	4	2	8	4
				2816	3520	2816	3520	2253
34	Cyanocobalamin injection	1000 µg /mL	0.30 EU/µg	1	2	1	4	2
				48000	60000	48000	60000	38400
35	Digoxin injection	0.25 mg/mL	200 EU/mg	16	32	16	8	32
				8000	10000	8000	10000	6400
36	Dimorpholamine injection3)	15 mg/mL	5.0 EU/mg	16	64	16	16	8
				12000	15000	12000	15000	9600
37	Water for injection†	—	0.25 EU/mL
38	Water for injection in containers	—	0.25 EU/mL	1	1	1	1	1
				—	—	—	—	—
39	Suxamethonium chloride injection	22 mg/mL	2.0 EU/mg	1	2	1	8	1
				7040	8800	7040	8800	5632
40	Suxamethonium chloride for injection	100 mg/mL	1.5 EU/mg	8	8	4	16	8
				24000	30000	24000	30000	19200
41	Streptomycin sulfate for injection	333 mg/mL	0.10 EU/mg	32	128	8	1024	16
				5333	6667	5333	6667	4267
42	Serum gonadotrophin†		0.1 EU/unit
43	Serum gonadotrophin for injection	200 unit/mL	0.1 EU/unit	164)	1	164)	1	164)
				3200	4000	3200	4000	2520
44	Human menopausal gonadotrophin	75 Unit5)/mL	0.66 EU/unit5,6)	164)	2	44)	1	44)
				7920	9900	7920	9900	6336
45	Human chorionic gonadotrophin†		0.03 EU/unit
46	Human chorionic gonadotrophin for injection	200 mg/mL	0.03 EU/unit	1	1	1	1	1
				960	1200	960	1200	768
47	Isotonic sodium chloride solution	—	0.50 EU/mL	1	1	1	2	1
				80	100	80	100	64
48	Cefazolin sodium hydrate†		0.10 EU/mg1)
49	Cefazolin sodium for injection	333 mg/mL	0.05 EU/mg1)	32	64	32	16	32
				2667	3333	2667	3333	2133
50	Cefepime dihydrochloride hydrate†		0.04 EU/mg1)
51	Cefepime dihydrochloride hydrate for injection	50 mg/mL	0.06 EU/mg1)	8	16	4	32	8
				480	600	480	600	384
52	Cefozopran hydrochloride†		0.05 EU/mg1)
53	Cefozopran hydrochloride for injection	100 mg/mL	0.05 EU/mg1)	16	64	8	128	16
				800	1000	800	1000	640
54	Cefotiam hydrochloride for injection	83.3 mg/mL	0.125 EU/mg	16	16	8	64	32
				1667	2083	1667	2083	1333
55	Ceftazidime for injection	200 mg/mL	0.067 EU/mg1)	32	64	8	64	16
				2144	2680	2144	2680	1715
56	Cefpirome sulfate	50 mg/mL	0.10 EU/mg	16	32	8	32	8
				800	1000	800	1000	640
57	Cefmetazole sodium for injection	400 mg/mL	0.06 EU/mg	32	32	64	64	64
				3840	4800	3840	4800	3072
58	Celmoleukin (genetic recombination)	40 Unit7)/mL	100 EU/mL	1	2‡	1	64	2
				16000	20000	16000	20000	12800
59	Tazobactam8)	450 mg/mL	0.04 EU/mg1)	128	128	64	64	128
				2880	3600	2880	3600	2304
60	Sodium bicarbonate injection	0.7 mg/mL	5.0 EU/mEq	4	4	4	8	8
				560	700	560	700	448
61	Thiamylal sodium for injection	25 mg/mL	1.0 EU/mg	32	16	16	16	16
				4000	5000	4000	5000	3200
62	Thiamine chloride hydrochloride injection	50 mg/mL	6.0 EU/mg	8	16‡	4	128	32
				48000	60000	48000	60000	38400
63	Thiopental sodium for injection	25 mg/mL	0.30 EU/mg	32	16	16	128	32
				1200	1500	1200	1500	960
64	Sodium thiosulfate hydrate	100 mg/mL	0.01 EU/mg	8	16	2	32	4
				160	200	160	200	128
65	Teicoplanin	66.7 mg/mL	0.75 EU/mg1)	512	1024‡	512	512	256
				8000	10000	8000	10000	6400
66	Dextran 40	0.1 g/mL	2.5 EU/g	1	2	1	4	4
				40	50	40	50	32
67	Dextran 40 injection9)	100 mg/mL	0.50 EU/mL	1	1	1	64	4
				80	100	80	100	64
68	Deslanoside injection	0.2 mg/mL	500 EU/mg	4	8	4	2	8
				16000	20000	16000	20000	12800
69	Teceleukin (genetic recombination)†		5 EU/mg10)
70	Teceleukin for injection (genetic recombination)	35 Unit7)/mL	5 EU/35 unit7)	2	2	2	8	2
				28000	35000	28000	35000	22400
71	Dehydrocholic acid injection	100 mg/mL	0.30 EU/mg	64	64	32	32	64
				4800	6000	4800	6000	3840
72	Doxorubicin hydrochloride for injection	10 mg/mL	2.50 EU/mg1)	256	512‡	32	32	64
				4000	5000	4000	5000	3200
73	Dopamine hydrochloride injection	20 mg/mL	4.2 EU/mg	8	4	2	4	4
				13440	16800	13440	16800	10752
74	Tobramycin injection	60 mg/mL	0.50 EU/mg1)	32	32‡	4	512	8
				4800	6000	4800	6000	3840
75	Tranexamic acid injection	50 mg/mL	0.12 EU/mg	2	4	2	2	2
				960	1200	960	1200	768
76	Nicardipine hydrochloride injection	1 mg/mL	8.33 EU/mg	16	16	16	8	8
				1333	1666	1333	1666	1066
77	Nicotinic acid injection	50 mg/mL	3.0 EU/mg	8	8	8	16	16
				24000	30000	24000	30000	19200
78	Neostigmine methylsulfate injection	0.5 mg/mL	5 EU/mg	1	1	1	2	1
				400	500	400	500	320
79	Noradrenaline injection	1 mg/mL	300 EU/mg	2	2	2	4	4
				48000	60000	48000	60000	38400
80	Sucrose	100 mg/mL	0.25 EU/mg	44)	1	84)	1	1
				4000	5000	4000	5000	3200
81	Vasopressin injection	20 unit/mL	15 EU/unit11)	4	32	4	32	4
				48000	60000	48000	60000	38400
82	Panipenem	25 mg/mL	0.15 EU/mg1)	16	16	8	8	8
				600	750	600	750	480
83	Papaverine hydrochloride injection	40 mg/mL	6.0 EU/mg	256	1024	64	128	64
				38400	48000	38400	48000	30720
84	Vancomycin hydrochloride injection	100 mg/mL	0.25 EU/mg	128	512	64	128	128
				4000	5000	4000	5000	3200
85	Hydralazine hydrochloride injection	20 mg/mL	5.0 EU/mg	16	16‡	8	32	16
				16000	20000	16000	20000	12800
86	Piperacillin hydrate†		0.07 EU/mg1)
87	Piperacillin sodium for injection	200 mg/mL	0.04 EU/mg1)	64	128	64	32	64
				1280	1600	1280	1600	1024
88	Pyridoxine hydrochloride injection	10 mg/mL	3.0 EU/mg	4	8	2	128	16
				4800	6000	4800	6000	3840
89	Vinblastine sulfate for injection	2 mg/mL	10 EU/mg	16	32	8	8	2
				3200	4000	3200	4000	2560
90	Famotidine injection	10 mg/mL	15 EU/mg	8	16‡	8	16	16
				24000	30000	24000	30000	19200
91	Famotidine for injection	10 mg/mL	15 EU/mg	4	8	2	16	8
				24000	30000	24000	30000	19200
92	Phenolsulfonphthalein injection	6 mg/mL	7.5 EU/mg	16	128‡	16	64	64
				80	100	80	100	64
93	Glucose injection	50 %	0.50 EU/mL	16	16	8	32	4
				80	100	80	100	64
94	Prednisolone succinate for injection	10 mg/mL	2.4 EU/mg12)	16	16	16	8	32
				3840	4800	3840	4800	3072
95	Procain hydrochloride injection	20 mg/mL	0.02 EU/mg	8	32	4	16	8
				64	80	64	80	51
96	Procainamide hydrochloride injection	100 mg/mL	0.30 EU/mg	32	64	16	32	32
				4800	6000	4800	6000	3840
97	Furosemide injection	10 mg/mL	1.25 EU/mg	4	4	8	8	8
				2000	2500	2000	2500	1600
98	Protamine sulfate injection	10 mg/mL	6.0 EU/mg	256	1024	4	4096	8
				9600	12000	9600	12000	7680
99	Flomoxef sodium for injection	250 mg/mL	0.025 EU/mg1)	32	64	16	32	32
				1000	1250	1000	1250	800
100	Pethidine hydrochloride injection	50 mg/mL	6.0 EU/mg	32	32‡	16	32	64
				48000	60000	48000	60000	38400
101	Heparin calcium	25000 Unit13)/mL	0.003 EU/unit13)	2000	8000	32	10000	64
				12000	15000	12000	15000	9600
102	Heparin sodium injection	1000 unit/mL	0.0030 EU/unit	64	128‡	2	512	2
				480	600	480	600	384
103	Peplomycin sulfate injection	1 mg/mL	1.5 EU/mg1)	2	2	4	1	4
				240	300	240	300	192
104	Benzylpenicillin potassium for injection	100000 unit/mL	0.000125 EU/unit	8	16	8	4	8
				2000	2500	2000	2500	1600
105	Fosfomycin sodium for injection	100 mg/mL	0.025 EU/mg1)	1	2	2	32	2
				400	500	400	500	320
106	Mitomycin c for injection	2 mg/mL	10 EU/mg1)	2	4	2	8	2
				3200	4000	3200	4000	2520
107	D-Mannitol injection	20 %	0.50 EU/mL	1	2	1	1	2
				80	100	80	100	64
108	Minocycline hydrochloride for injection	50 mg/mL	1.25 EU/mg1)	2000	8000	512	2048	128
				10000	12500	10000	12500	8000
109	Mepivacain hydrochlorideInjection	20 mg/mL	0.6 EU/mg	8	32	4	8	8
				1920	2400	1920	2400	1536
110	Meropenem for injection	50 mg/mL	0.12 EU/mg1)	16	16	8	8	16
				960	1200	960	1200	768
111	Morphine hydrochloride injection	10 mg/mL	1.5 EU/mg	4	8‡	2	16	4
				2400	3000	2400	3000	1920
112	Lidocain injection	20 mg/mL	1.0 EU/mg	8	32	4	8	4
				3200	4000	3200	4000	2560
113	Riboflavin sodium phosphate injection	10 mg/mL	10 EU/mg	16	8	16	32	8
				16000	20000	16000	20000	12800
114	Magnesium sulfate injection	60.2 mg/mL	0.09 EU/mg	8	32	1	16	2
				867	1084	867	1084	694
115	Ringer’s solution	—	0.50 EU/mL	1	1	1	2	1
				80	100	80	100	64
116	Lincomycin hydrochloride injection	300 mg/mL	0.50 EU/mg1)	32	128	32	64	32
				24000	30000	24000	30000	19200
117	Levallorphan tartrate injection	1 mg/mL	150 EU/mg	2	8	2	16	2
				24000	30000	24000	30000	19200
118	Roxatidine acetate hydrochloride for injection	15 mg/mL	4.0 EU/mg	4	32	2	16	4
				9600	12000	9600	12000	7680

1) Potency; 2) Insulin unit; 3) Performed the test at 0.15 (w/v) %; 4) Because of its contamination, the sample was diluted to the extent where the measured value entered the range of the standard curve; 5) Follicle-stimulating hormone unit; 6) Performed the test by dissolving 75 unit of follicle-stimulating hormone per mL of water for Bacterial Endotoxins Test; 7) × 10000 units; 8) Tazobactam sodium/piperacillin sodium for injection (reference); 9) Low-molecular-weight dextran-added lactated Ringer’s solution; 10) Per 1 mg of protein;11) Vasopressin unit; 12) Amounts corresponding to prednisolone; 13) Heparin unit; MVD, maximum valid dilution. ^†: Drug substance that was not available for us. ^‡: 60-min incubation.

Comparison between PyroSmart and Limulus Lysate Reagents

Potencies of endotoxins—The potencies of three typical types of endotoxins were measured with the three recombinant reagents including PyroSmart and six lysate reagents produced from the horseshoe crab (Fig. 4). There was no significant difference based on F-test and t-test (p = 0.05) between the recombinant reagents (generating four data sets) and lysate reagents (generating six data sets).

Fig. 4. Comparison of the Reactivity of Three Kinds of Endotoxins in the Assay with Recombinant Reagents and Lysate Reagents

Reactivity is expressed as EU/ng using USP reference standard endotoxin (RSE) as a reference. Error bars indicate standard deviation in the results of three lots. Two lots of EndoZyme and one lot of Endozyme II were used for the study.

Figure 4 also shows variations in the potencies as error bars with each reagent between the three lots. In a few cases lot-to-lot variances (CV; >20%) in the potencies were recognized; P. aeruginosa 10 with Endochrome-K and S. minnesota R595 Re with PyroGene and Endochrome-K. In the case of EndoZyme, only two lots were available and one lot of Endozyme II was used instead.

Recoveries of endotoxin added to parenteral drugs—The recoveries of endotoxins added into twenty seven parenteral drugs, which were evenly selected according to the wide range of susceptibility to interfering factors, were measured with three kinds of recombinant reagents and five different lysate reagents. The minimum dilution factors, NIDs, are shown in Table 4.

Table 4. Non-interfering-Dilution (NID) of Injectable Drugs Tested with Recombinant Reagents and Lysate Reagents

(A) Recombinant reagents
ID	Parenteral drugs	NID (upper line) / MVD (lower line)
		PyroSmart						PyroGene			EndoZyme
		Rate			Onset time			Endpoint			Endpoint
		Lot
		1st.	2nd.	3rd.	1st.	2nd.	3rd.	1st.	2nd.	3rd.	1st.		2nd.		3rd.
1	Aminophylline injection	16	16	16	32	16	32	4	8	8	Not tested
		2400			3000			3000			3000
2	Alprostadil injection	8	8	8	256*	128*	128*	256	512	256	128		128		128
		1600			2000			2000			2000
3	Idarubicin hydrochloride for injection	256	256	256	512	512	512	128	128	128	32		32		32
		1424			1780			1780			1780
4	Calcium chloride injection	16	16	16	32	64	32	128	128	64	128		64		128
		2664			3330			3330			3330
5	10% Sodium chloride injection	2	2	2	4	4	4	8	4	8	4		4		4
		576			720			720			720
6	Xylitol injection	1	1	1	1	1	1	1	1	1	1		1		1
		80			100			100			100
7	Sodium citrate injection for transfusion	2	2	1	1	1	1	64	64	128	128		128		128
		896			1120			1120			1120
8	Clindamycin phosphate injection	128	128	128	64	128	64	16	16	16	32		32		32
		2400			3000			3000			3000
9	Digoxin injection	16	16	16	16	16	16	8	8	4	8		8		8
		8000			10000			10000			10000
10	Dimorpholamine injection	16	16	16	32*	16*	32*	8	4	8	8		8		8
		12000			15000			15000			15000
11	Isotonic sodium chloride solution	1	1	1	1	1	1	1	1	1	1		1		1
		80			100			100			100
12	Cefazolin sodium for injection	32	32	32	128	64	64	128	256	128	64		64		64
		2667			3333			3333			3333
13	Cefmetazole sodium for injection	32	32	32	64	32	64	64	64	64	64		64		64
		3840			4800			4800			4800
14	Tazobactam	128	128	128	256	128	256	64	64	64	64		64		64
		2880			3600			3600			3600
15	Sodium bicarbonate injection	8	8	8	8	4	8	16	16	16	8		8		8
		560			700			700			700
16	Teicoplanin	512	512	512	1024*	1024*	1024*	1024	1024	2048	2048		2048		1024
		8000			10000			10000			10000
17	Dehydrocholic acid injection	64	32	64	64	64	64	32	32	32	32		64		32
		4800			6000			6000			6000
18	Doxorubicin hydrochloride for injection	1024	512	1024	512*	512*	512*	512	512	512	256		256		256
		4000			5000			5000			5000
19	Nicardipine hydrochloride injection	16	16	16	16	16	32	64	64	64	32		32		32
		1333			1666			1666			1666
20	Nicotinic acid injection	32	64	64	32	32	64	16	16	16	8		16		16
		24000			30000			30000			30000
21	Papaverine hydrochloride injection	256	256	256	1024	512	512	256	128	128	256		256		256
		38400			48000			48000			48000
22	Vancomycin hydrochloride injection	256	64	128	512	256	512	64	128	128	64		64		64
		4000			5000			5000			5000
23	Piperacillin sodium for injection	64	64	64	128	128	128	32	32	32	32		32		64
		1280			1600			1600			1600
24	Famotidine injection	8	8	8	16	16	8	4	4	4	16		16		8
		24000			30000			30000			30000
25	Glucose injection	8	4	4	8	8	16	4	4	4	4		8		8
		80			100			100			100
26	Heparin calcium	1000	2000	1000	4000*	1000*	2000*	>30000	>30000	>30000	>30000		>30000		>30000
		12000			15000			15000			15000
27	D-Mannitol injection	1	1	1	2	1	1	2	2	1	1		2		2
		80			100			100			100
(B) Lysate reagents
ID	Parenteral drugs	NID (upper line)/MVD (lower line)
		Endospecy ES-50M			Pyrochrome			Kinetic-QCL			Limulus ES-II			Limulus color KY test
		Rate			Onset time			Onset time			Onset time			Onset time
		Lot
		1st.	2nd.	3rd.	1st.	2nd.	3rd.	1st.	2nd.	3rd.	1st.	2nd.	3rd.	1st.	2nd.	3rd.
1	Aminophylline injection	8	8	8	16	8	8	4	4	4	8	8	8	8	8	8
		2400			3000			3000			1920			30000
2	Alprostadil injection	16	16	16	32	256	128	4	8	8	256	256	256	256	256	256
		1600			2000			2000			1280			20000
3	Idarubicin hydrochloride for injection	64	256	64	128	64	128	32	32	32	32	512	512	32	32	128
		1424			1780			1780			1139			17800
4	Calcium chloride injection	32	64	32	16	16	16	256	128	128	4	16	16	8	8	8
		2664			3330			3330			2131			33300
5	10% Sodium chloride injection	4	4	4	2	2	2	16	16	16	4	2	4	1	1	1
		576			720			720			461			7200
6	Xylitol injection	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1
		80			100			100			64			1000
7	Sodium citrate injection for transfusion	4	8	8	8	8	8	64	128	64	4	4	4	1	1	1
		896			1120			1120			717			11200
8	Clindamycin phosphate injection	32	32	32	16	16	16	32	32	32	32	64	32	64	64	64
		2400			3000			3000			1920			30000
9	Digoxin injection	16	16	16	16	16	16	8	8	8	16	16	16	16	16	16
		8000			10000			10000			6400			100000
10	Dimorpholamine injection	8	8	8	16	8	16	16	16	16	8	8	16	4	8	16
		12000			15000			15000			9600			150000
11	Isotonic sodium chloride solution	1	1	1	1	1	1	1	2	2	1	1	1	1	1	1
		80			100			100			64			1000
12	Cefazolin sodium for injection	32	32	32	16	16	16	128	128	64	32	32	32	32	16	16
		2667			3333			3333			2133			33330
13	Cefmetazole sodium for injection	64	64	64	64	64	64	64	128	128	64	64	16	32	32	32
		3840			4800			4800			3072			48000
14	Tazobactam	128	128	128	256	64	64	64	64	64	64	128	64	32	32	32
		2880			3600			3600			2304			36000
15	Sodium bicarbonate injection	4	8	4	4	4	8	8	16	16	8	8	8	8	4	4
		560			700			700			448			7000
16	Teicoplanin	512	512	512	1024	512	512	1024	512	512	256	256	256	1024	1024	1024
		8000			10000			10000			6400			100000
17	Dehydrocholic acid injection	32	32	32	64	32	32	32	32	32	64	64	64	32	32	32
		4800			6000			6000			3840			60000
18	Doxorubicin hydrochloride for injection	64	128	4	2048	16	4	128	64	64	32	2048	1024	128	16	32
		4000			5000			5000			3200			50000
19	Nicardipine hydrochloride injection	8	16	16	32	32	32	16	8	8	16	128	64	16	32	32
		1333			1666			1666			1066			16660
20	Nicotinic acid injection	32	64	64	16	16	64	16	16	16	16	32	16	64	64	32
		24000			30000			30000			19200			300000
21	Papaverine hydrochloride injection	64	64	64	512	256	512	128	128	128	64	1024	512	64	128	128
		38400			48000			48000			30720			480000
22	Vancomycin hydrochloride injection	32	64	32	128	256	256	128	128	128	64	128	128	64	64	64
		4000			5000			5000			3200			50000
23	Piperacillin sodium for injection	64	128	64	64	32	64	32	32	64	64	64	64	16	32	32
		1280			1600			1600			1024			16000
24	Famotidine injection	8	4	4	16	8	8	8	8	8	16	8	8	16	2	8
		24000			30000			30000			19200			300000
25	Glucose injection	4	8	4	8	8	8	4	8	8	4	4	4	4	4	4
		80			100			100			64			1000
26	Heparin calcium	64	64	64	8	8	8	10000	8000	8000	16	8	8	16	1024	1024
		12000			15000			15000			9600			150000
27	D-Mannitol injection	2	2	1	2	1	1	2	2	2	2	2	1	1	1	1
		80			100			100			64			1000

* Sixty minutes assay was performed. Shadows indicate that the difference in the NID among 3 lots is equal to or more than eight-fold.

The variation in the NIDs within the three recombinant reagents is similar to that within five lysate reagents. All three recombinant reagents show little lot-to-lot variations of endotoxin-recoveries in all twenty-seven drugs. In contrast, four out of five lysate reagents showed lot-to-lot variations equal to or larger than 8-fold in seven of twenty-seven drugs examined.

DISCUSSION

As reported in literature,^22,27–32) endotoxins from different bacterial strains showed highly varying potencies. This is due to their differing structures of the lipid A moiety and polysaccharide length (1.87 EU/ng of E. coli J5 versus 118 EU/ng of E. coli F583 Rd2) (Fig. 1).

The variance of the potency of each endotoxin among lysate reagents is maximally 5.5-fold and minimally 1.7-fold. On the other hand, all potencies of the thirteen endotoxins when tested with PyroSmart fell in the range between 50 of minimum and 200% of maximum of those with the three lysate reagents. In addition, potencies except for that of S. typhimurium obtained with PyroSmart fell in the range of the mean+/−3SD of the potencies with the three lysate reagents, suggesting again that the reactivity of PyroSmart to each endotoxin is almost equal to those of the three lysate reagents. Figure 4 also shows no significant differences in reactivities of the three recombinant reagents and the six lysate reagents.

All analytical characteristics in method validation of the PyroSmart satisfied the acceptance criteria (Table 2). Thus, the recombinant chromogenic reagent, PyroSmart can be applicable as an alternative to the BET in the pharmacopeias.

All minimum dilution factors expressed as NIDs of the 109 injectable drugs measured with PyroSmart were smaller than the MVDs of respective drugs (Table 3). In further experiment, the three recombinant reagents showed a similar susceptibility to interference in 27 drugs as the five lysate reagents (Table 4). Moreover, the lot-to-lot variation of susceptibility in the assays with the recombinant reagents was equal to or smaller than those with lysate reagents. Of the three recombinant reagents, only PyroSmart was able to detect endotoxin in the drug, Heparin Calcium, which strongly inhibited the detection by other recombinant reagents. The difference in susceptibility to the interference derived from Heparin Calcium between PyroSmart and the two recombinant factor C reagents may be attributable to the components, other than factor C, contained in PyroSmart. Heparin is known to interact with factors C and B,^33,34) therefore Heparin Calcium may reduce the quantity of recombinant factor C available for endotoxin in PyroGene and EndoZyme also in PyroSmart. However factor B is also reported to be involved in endotoxin recognition as well as in signal transduction from the upstream factor C.³⁵⁾ Thus, the presence of factor B in PyroSmart may mitigate the strong inhibition by Heparin Calcium.

It is also reported that the recombinant factor C in PyroSmart produced from mammalian cells is less susceptible to interference by some substances than the recombinant products only containing factor C from insect cells or factor C purified from horseshoe crab.²¹⁾

In conclusion, an endotoxin assay using recombinant reagents can be an effective alternative assay for the Bacterial Endotoxins Test in the pharmacopeias. It would resolve the inconsistencies in test results caused by lot-to-lot variance of the lysate reagents and contribute to the conservation of the horseshoe crab.

Acknowledgments

This work was supported in part by a grant “Study of pharmaceutical and medical device regulatory science” from the Health Labour Sciences Research Grant and “Research on Regulatory Harmonization and Evaluation of Pharmaceuticals, Medical Devices, Regenerative and Cellular Therapy Products, Gene Therapy Products, and Cosmetics” from the Japan Agency for Medical Research and Development, AMED.

Conflict of Interest

Ogura, Mizumura, Aketagawa and Oda are employees of Seikagaku Corporation. Muroi and Tanamoto declare no conflict of interest.

Supplementary Materials

The online version of this article contains supplementary materials.

REFERENCES

• 1) Schletter J, Heine H, Ulmer AJ, Rietschel ET. Molecular mechanisms of endotoxin activity. Arch. Microbiol., 164, 383–389 (1995).
• 2) Galanos C, Freudenberg MA. Bacterial endotoxins: biological properties and mechanisms of action. Mediators Inflamm., 2, S11–S16 (1993).
• 3) Alexander C, Rietschel ET. Bacterial lipopolysaccharides and innate immunity. J. Endotoxin Res., 7, 167–202 (2001).
• 4) Ramachandran G. Gram-positive and Gram-negative bacterial toxins in sepsis: a brief review. Virulence, 5, 213–218 (2014).
• 5) Fox CB, Friede M, Reed SG, Ireton GC. Synthetic and natural TLR4 agonists as safe and effective vaccine adjuvants. Subcell. Biochem., 53, 303–321 (2010).
• 6) Levin J, Bang FB. The role of endotoxin in the extracellular coagulation of limulus blood. Bull. Johns Hopkins Hosp., 115, 265–274 (1964).
• 7) Levin J, Hochstein HD, Novitsky TJ. Clotting cells and Limulus Amoebocyte lysate: an amazing analytical tool. The American horseshoe crab. (Shuster CNJ, Barlow RB, Brockman HJ eds.) Harvard University Press, Cambridge, pp. 310–340 (2003).
• 8) Tanaka S, Iwanaga S. Limulus test for detecting bacterial endotoxins. Methods Enzymol., 223, 358–364 (1993).
• 9) Kawabata S, Muta T. Sadaaki Iwanaga: discovery of the lipopolysaccharide- and β-1,3-D-glucan-mediated proteolytic cascade and unique proteins in invertebrate immunity. J. Biochem., 147, 611–618 (2010).
• 10) Japanese Pharmacopoeia. 4.01 Bacterial endotoxins test.
• 11) United States Pharmacopeia. Chapter ‹85› Bacterial endotoxins test.
• 12) European Pharmacopoeia. 2.6.14. Bacterial endotoxins.
• 13) Anderson RL, Watson WH 3rd, Chabot CC. Sublethal behavioral and physiological effects of the biomedical bleeding process on the American horseshoe crab, Limulus polyphemus. Biol. Bull., 225, 137–151 (2013).
• 14) Botton ML, Carmichael RH, Shin PKS. Emerging issues in horseshoe crab conservation: a perspective from the IUCN species specialist group. Global perspectives on horseshoe crab biology, conservation and management. (Carmichael RH, Botton ML, Shin PKS, Changing SGC eds.) Springer International Publishing, Cham, Switzerland, pp. 369–381 (2015).
• 15) Lee CN, Morton B. Changes in the distributions of juvenile horseshoe crabs (Arthropoda: Chelicerata) (2002–2014) related to environmental perturbations at Pak Nai and Ha Pak Nai, Deep Bay, Hong Kong SAR, China. Mar. Pollut. Bull., 108, 134–146 (2016).
• 16) Twohy CW, Nierman ML, Duran AP, Munson TE. Comparison of limulus amoebocyte lysates from different manufacturers. J. Parenter. Sci. Technol., 37, 93–96 (1983).
• 17) Milton DK, Johnson DK, Park JH. Environmental endotoxin measurement: interference and sources of variation in the Limulus assay of house dust. Am. Ind. Hyg. Assoc. J., 58, 861–867 (1997).
• 18) McKenzie JH, Alwis KU, Sordillo JE, Kalluri KS, Milton DK. Evaluation of lot-to-lot repeatability and effect of assay media choice in the recombinant Factor C assay. J. Environ. Monit., 13, 1739–1745 (2011).
• 19) Ding JL, Ho B. Endotoxin detection—from limulus amoebocyte lysate to recombinant factor C. Subcell. Biochem., 53, 187–208 (2010).
• 20) Bolden JS, Smith KR. Application of recombinant Factor C reagent for the detection of bacterial endotoxins in pharmaceutical products. PDA J. Pharm. Sci. Technol., 71, 405–412 (2017).
• 21) Mizumura H, Ogura N, Aketagawa J, Aizawa M, Kobayashi Y, Kawabata S, Oda T. Genetic engineering approach to develop next-generation reagents for endotoxin quantification. Innate Immunity, 23, 136–146 (2017).
• 22) Abate W, Sattar AA, Liu J, Conway ME, Jackson SK. Evaluation of recombinant factor C assay for the detection of divergent lipopolysaccharide structural species and comparison with Limulus amoebocyte lysate-based assays and a human monocyte activity assay. J. Med. Microbiol., 66, 888–897 (2017).
• 23) Loverock B, Simon B, Burgenson A, Baines A. A recombinant factor C procedure for the detection of Gram-negative bacterial endotoxin. Pharmacop. Forum, 36, 321–329 (2010).
• 24) International conference of harmonization of technical requirements for registration of pharmaceuticals for human use. ICH harmonized tripartite guideline. Validation of analytical procedures: text and methodology Q2 (R1), November 2005.
• 25) Pharmaceutical Safety and Environmental Health Bureau, Ministry of Health, Labour and Welfare (Japan): Guideline on Bioanalytical Method (Ligand Binding Assay) Validation in Pharmaceutical Development, May 2014.
• 26) Saito H, Misaki A, Harada T. A comparison of the structure of curdlan and pachyman. Agric. Biol. Chem., 32, 1261–1269 (1968).
• 27) Tanamoto K, Abe C, Homma JY, Kojima Y. Regions of the lipopolysaccharide of Pseudomonas aeruginosa essential for antitumor and interferon-inducing activities. Eur. J. Biochem., 97, 623–629 (1979).
• 28) Tanamoto K, Zähringer U, McKenzie GR, Galanos C, Rietschel ET, Lüderitz O, Kusumoto S, Shiba T. Biological activities of synthetic lipid A analogs: pyrogenicity, lethal toxicity, anticomplement activity, and induction of gelation of Limulus amoebocyte lysate. Infect. Immun., 44, 421–426 (1984).
• 29) Proctor RA, Textor JA. Activation and inhibition of Limulus amoebocyte lysate coagulation by chemically defined substructures of lipid A. Infect. Immun., 49, 286–290 (1985).
• 30) Kanegasaki S, Tanamoto K, Yasuda T, Homma JY, Matsuura M, Nakatsuka M, Kumazawa Y, Yamamoto A, Shiba T, Kusumoto S, Imoto M, Yoshimura H, Shimamoto T. Structure–activity relationship of lipid A: comparison of biological activities of natural and synthetic lipid A’s with different fatty acid compositions. J. Biochem., 99, 1203–1210 (1986).
• 31) Rietschel ET, Kirikae T, Schade FU, Mamat U, Schmidt G, Loppnow H, Ulmer AJ, Zähringer U, Seydel U, Di Padova F. Bacterial endotoxin: molecular relationships of structure to activity and function. FASEB J., 8, 217–225 (1994).
• 32) Takada H, Kotani S, Tanaka S, Ogawa T, Takahashi I, Tsujimoto M, Komuro T, Shiba T, Kusumoto S, Kusunose N, Hasegawa A, Kiso M. Structural requirements of lipid A species in activation of clotting enzymes from the horseshoe crab, and the human complement cascade. Eur. J. Biochem., 175, 573–580 (1988).
• 33) Morita T, Tanaka S, Nakamura T, Iwanaga S. A new (1 → 3)-β-D-glucan-mediated coagulation pathway found in limulus amebocytes. FEBS Lett., 129, 318–321 (1981).
• 34) Nakamura S, Levin J. Fractionation of Limulus amebocyte lysate: characterization of activation of the proclotting enzyme by an endotoxin-mediated activator. Biochim. Biophys. Acta, 707, 217–225 (1982).
• 35) Kobayashi Y, Takahashi T, Shibata T, Ikeda S, Koshiba T, Mizumura H, Oda T, Kawabata S. Factor B is the second lipopolysaccharide-binding protease zymogen in the horseshoe crab coagulation cascade. J. Biol. Chem., 290, 19379–19386 (2015).