Biologicals 44 (2016) 417e422

Contents lists available at ScienceDirect

Biologicals
journal homepage: www.elsevier.com/locate/biologicals

Masking of endotoxin in surfactant samples: Effects on Limulus-based

detection systems
Johannes Reich a, *, Pierre Lang b, 1, Holger Grallert c, Hubert Motschmann a
a
Institute of Physical and Theoretical Chemistry, University of Regensburg, Regensburg, Germany
b
Quality Control & Assurance, Pharmaceuticals Division, F. Hoffmann-La Roche Ltd, Basel, Switzerland
c
Research and Development, Hyglos GmbH, Bernried, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Over the last few decades Limulus Amebocyte Lysate (LAL) has been the most sensitive method for the
Received 11 September 2015 detection of endotoxins (Lipopolysaccharides) and is well accepted in a broad field of applications.
Received in revised form Recently, Low Endotoxin Recovery (LER) in biopharmaceutical drug products has been noticed, whereby
24 March 2016
the detection of potential endotoxin contaminations is not ensured. Notably, most of these drug products
Accepted 25 April 2016
Available online 25 July 2016
contain surfactants, which can have crucial effects on the detectability of endotoxin. In order to analyze
the driving forces of LER, endotoxin detection in samples containing nonionic surfactants in various
buffer systems was investigated. The results show that the process of LER is kinetically controlled and
Keywords:
Endotoxin
temperature-dependent. Furthermore, only the simultaneous presence of nonionic surfactants and
Lipopolysaccharide components capable of forming metal complexes resulted in LER. In addition, capacity experiments show
Low endotoxin recovery that even hazardous amounts of endotoxin can remain undetectable within such formulation compo-
LER sitions. In conclusion, the LER phenomenon is caused by endotoxin masking and not by test interference.
Limulus amebocyte lysate In this process, the supramolecular structure of endotoxin is altered and exhibits only a limited sus-
LAL ceptibility in binding to the Factor C of Limulus-based detection systems. We propose a two-step
mechanism of endotoxin masking by complex forming agents and nonionic surfactants.
© 2016 The Author(s). Published by Elsevier Ltd on behalf of International Alliance for Biological
Standardization. This is an open access article under the CC BY license (http://creativecommons.org/
licenses/by/4.0/).

1. Introduction One reason for inadequate detection of endotoxin is interference

of sample constituents with the enzymatic reaction of the Limulus-
Endotoxins are products of Gram-negative bacteria, released based detection system. In this case, certain components (e.g. heavy
during bacterial cell division, lysis and at cell death. Chemically, metals, protease inhibitors) can directly disturb enzyme activation of
endotoxins are lipopolysaccharides (LPS). The general structure of the detection system, which is called test interference [6]. This
LPS comprises three basic units (Fig. 1), O-antigen, Core and Lipid A, phenomenon is well known and to indicate test interference, posi-
wherein the latter is the highly toxic part [1]. Within the tive product controls are performed. To this end, a known amount of
mammalian blood circulation, endotoxin can trigger severe physi- endotoxin is added to the sample and immediately measured. A test
ological reactions (e.g. fever, septic shock) [2,3]. Thus, bacterial is considered valid if the spiked endotoxin is recovered in a range of
endotoxin testing of drug products for parenteral administration is 50e200%. If the validity criterion is not fulfilled, it is recommended to
mandatory. However, several applications have shown that detec- overcome interference by suitable sample treatments such as dilu-
tion of endotoxin using widespread Limulus Amebocyte Lysate tion, filtration, neutralization, dialysis or heating etc. [7]. Another
(LAL) based methods are not always feasible in complex protein potential reason for inadequate endotoxin detection is the interac-
samples containing endotoxin [4,5]. tion of endotoxin itself with matrix components of the sample. For
instance, it has been reported that endotoxin can interact with blood
components [8], proteins [4] or amphiphilic molecules [9,10],
resulting in a significant change of endotoxin activity. Notably, ap-
* Corresponding author. An der Baerenmuehle 6, 82362 Weilheim, Germany. proaches which eliminate test interference problems are not effec-
Tel.: þ49 8158 9060 0; fax: þ49 8158 9060 210. tive in overcoming such effects [4]. In the 1990's Greaves and co-
E-mail address: [email protected] (J. Reich). workers [11] already differentiated between dilution dependent
1
Retired since June 2015.

http://dx.doi.org/10.1016/j.biologicals.2016.04.012
1045-1056/© 2016 The Author(s). Published by Elsevier Ltd on behalf of International Alliance for Biological Standardization. This is an open access article under the CC BY
license (http://creativecommons.org/licenses/by/4.0/).
418 J. Reich et al. / Biologicals 44 (2016) 417e422

(LPS) out of a 10,000 EU/mL stock solution. The pH of buffer com-

ponents was adjusted to 7.5, if not otherwise specified. Before
adding the spikes to the sample, the stock solution was shaken at
1400 rpm for 10 min. After spiking, the samples were incubated at
defined temperatures and periods of time. Immediately after in-
cubation, the samples were mixed at 1400 rpm for 2 min again, and
diluted in depyrogenated water in order to avoid test interference.
Necessary dilutions of the particular sample compositions were
determined prior to the actual experiment.
Kinetics: All samples with different incubation periods were
measured on the same microtiter plate, to avoid variation from test
to test. Therefore, endotoxin recovery kinetics was performed in a
reverse manner. The particular sample was aliquoted and all ali-
quots were stored under equal conditions over time. The aliquot
with the longest endotoxin incubation period was spiked first (e.g.
7 days prior to the measurement). Further aliquots with shorter
incubation periods were spiked later in accordance with the
respective incubation period. The zero time point aliquot was
spiked immediately before measurement. To control accuracy of
the spikes at different time points, equal amounts of endotoxin
Fig. 1. Schematic structure of Lipopolysaccharides (LPS). LPS is an amphiphilic mole-
were spiked into depyrogenated water (data not shown).
cule. The fatty acids within the Lipid A are hydrophobic and the polysaccharides in the Detection of endotoxin: For detection of endotoxin, a recombi-
Core and O-antigen are hydrophilic. In addition, LPS is electrically charged due to nant Factor C test (EndoZyme®) and a kinetic chromogenic Limulus
substitution (e.g. phosphates) in the core region and on the diglucosamine of Lipid A. Amebocyte Lysate (LAL) test were used according to manufacturer's
With regard to the biological nature, LPS can be divided into the three functional
instructions. The released amount of fluorescence substrate, using
subunits O-antigen, Core and Lipid A. The latter is the toxic fragment of the molecule.
the recombinant test, was measured spectrophotometrically at
440 nm with an FLx800 fluorescence microplate reader (BioTek,
interference and dilution-independent interference in environ- Bad Friedrichshall, Germany). All samples were measured in
mental samples. The latter phenomenon is called masking. duplicate and average values were used for further calculations.
In the recent past, inadequate endotoxin recovery over time has Endotoxin concentrations (EU/mL) were calculated using Gen5
been observed in biopharmaceutical drug products [12]. In such Data Analysis Software Version 2.05 (BioTek, Bad Friedrichshall,
cases, the active pharmaceutical ingredients are mostly proteins Germany). Standard curves were fit using a four parameter logistic
[13], which are capable of intrinsic binding to endotoxin as previ- non-linear regression model. The detection limit of the assay was
ously described by Anspach and co-workers [4]. The inadequate 0.005 EU/mL. For the LAL test, the released amount of chromogenic
detection of endotoxin might be explained by protein-endotoxin substrate was measured spectrophotometrically at 405 nm with an
interactions. Nonetheless, therapeutic proteins are usually stabi- Epoch2 absorbance microplate reader (BioTek, Bad Friedrichshall,
lized by excipients, like nonionic surfactants and certain buffer Germany). All samples were measured in duplicate and average
components [14]. Surprisingly, endotoxin spiking experiments in values were used for further calculations. Endotoxin concentrations
formulations that lack the active pharmaceutical ingredient (e.g. (EU/mL) were calculated using Gen5 Data Analysis Software
monoclonal antibody) resulted in endotoxin masking over time. Version 2.05 (BioTek, Bad Friedrichshall, Germany). Standard curves
Such observations of disturbed endotoxin determinations in bio- were fit using a linear regression model. Detection limit of the assay
pharmaceutical products over time and the related risk of undis- was 0.005 EU/mL.
covered endotoxin contamination events compelled us to study the Calculation of endotoxin recovery [%]: The determined endotoxin
impact of common formulation components on the detectability in concentrations in the tested samples were compared to the endo-
Limulus-based detection systems. The aim of the present study is to toxin concentrations at time zero in positive controls and stated as
understand the mechanism of Low Endotoxin Recovery (LER) in percent. Positive controls were prepared by spiking LPS into
samples containing nonionic surfactants in combination with depyrogenated water.
standard buffer systems.
3. Results
2. Materials & methods
LER was observed in various samples. In the beginning of the
E.coli O55:B5 lipopolysaccharide (gel-filtered), polysorbate 20, study crucial formulation components of common bio-
polysorbate 80, octoxynol 9, citric acid, trisodium citrate, phos- pharmaceuticals were examined. Therefore, endotoxin masking of
phoric acid, sodium dihydrogen phosphate, disodium hydrogen single and multiple components were investigated. The end-point
phosphate were obtained from SigmaeAldrich Chemicals, Stein- of the reaction was determined by endotoxin recovery kinetics at
heim, Germany. Depyrogenated water, depyrogenated borosilicate different temperatures. While multi-parameter interactions be-
glass tubes and recombinant Factor C test (EndoZyme®) were ob- tween surfactants, complex forming agents and endotoxin were
tained from Hyglos GmbH, Bernried, Germany. Kinetic chromo- observed, the focus of the investigation was the particular impact of
genic Limulus Amebocyte Lysate test was obtained from Lonza Inc., these components on the detectability of endotoxin. Thus, the
Walkersville, USA. Prior to the experiments, all relevant materials impact of pH and different buffer systems as well as the effects of
were tested for endotoxin content and proven to contain less than different nonionic surfactants were studied. Finally, various endo-
0.005 EU/mL. toxin concentrations were added to a LER causing formulation to
Sample preparation: Samples were prepared in 5 mL glass tubes evaluate the masking capacity.
with sample volumes of 1 mL per sample. Unless otherwise Single and mixtures of common formulation components were
described, samples were spiked with 10 mL of lipopolysaccharide examined to identify critical components or component
 J. Reich et al. / Biologicals 44 (2016) 417e422 419

Table 1 additions showed no significant loss of activity over time, according

Endotoxin recovery over time in presence of single and multiple formulation to the validity criteria of 50%e200% of endotoxin recovery. In
components.
contrast, endotoxin could not be measured in samples containing
Sample Formulation components: T0 recovery [%] T7 recovery [%] both buffer and surfactant (samples 5 and 6) after an incubation
1 H2O 100 94 period of seven days. Therefore, the kinetics of endotoxin recovery
2 Sodium citrate 125 94 in samples containing polysorbate 20 and sodium citrate was
3 Sodium phosphate 95 69 analyzed. Fig. 2 shows the endotoxin recovery of three identical
4 Polysorbate 20 91 79
samples as a function of time at different incubation temperatures
5 Sodium citrate þ 1 0
polysorbate 20 (4  C, RT and 37  C) using a LAL test (A) and a recombinant Factor C
6 Sodium phosphate þ 52 0 test (B) for detection. After a certain period of incubation, all sam-
polysorbate 20 ples showed low endotoxin recovery in both detection systems.
Samples were spiked with an endotoxin amount of 10,000 EU/mL. Endotoxin was This result clearly indicates that this phenomenon is independent
detected after preparation (approx. 45 min., T0) and after sample incubation of of the test system. Furthermore, the loss of activity was significantly
seven days (T7) at room temperature. Prior to the measurement samples were accelerated with increasing incubation temperature.
diluted (1:1000 and 1:10,000).
The latter experiments show that only the combination of a
combinations affecting endotoxin detection. Endotoxin recovery in buffer system and a surfactant results in LER. Thus, the impact of
the presence of different formulation components are shown in different buffer systems was studied and is shown in Fig. 3. In order
Table 1. The recovery was compared directly after sample prepa- to investigate pH dependency of endotoxin recovery over time,
ration (T0) and after sample incubation for seven days (T7) at room different pH conditions were studied (Fig. 3A). In the absence of
temperature. Samples 2, 3 and 4, containing only single component surfactants, the variation of pH had no impact on endotoxin

Fig. 2. Endotoxin recovery kinetics in citrate-polysorbate formulations. 100 EU/mL endotoxin were added to samples containing 10 mM citrate and 0.05% polysorbate 20 and
incubated for different time periods. The endotoxin recovery is plotted as a function of the incubation time. The different curves indicate incubation temperatures at 36e38  C (C),
21e23  C (:) and 2e8  C (-). For detection kinetic chromogenic LAL test (A) and recombinant Factor C tests (B) were used. The error bars reflect the standard deviation of three
independent replicates (n ¼ 3) of the sample. The replicates were measured on the same microtiter plate.

Fig. 3. Impact of buffer system on endotoxin recovery. (A): Effect of pH on endotoxin recovery: 100 EU/mL of endotoxin was added to solutions containing 0.05 wt % polysorbate 20
▫ ⋄
plus 10 mM citrate (-), 0.05 wt % polysorbate 20 plus 10 mM phosphate (A), citrate only ( ) and phosphate only ( ). The pH varied in a range from 1 to 9 and incubation was at
room temperature for seven days. The endotoxin recovery is shown as a function of the pH. (B): Effect of buffer systems on kinetics of endotoxin recovery: 100 EU/mL of endotoxin
were added to solutions containing a buffer (5 mM EDTA (:), 10 mM sodium citrate (A) or 10 mM sodium phosphate (-)) and 0.05 wt % polysorbate 20. Sample incubation was at
room temperature. The endotoxin recovery is plotted as function of the incubation time. The error bars reflect the standard deviation of three independent replicates (n ¼ 3) of the
sample. The replicates were measured on the same microtiter plate.
420 J. Reich et al. / Biologicals 44 (2016) 417e422

detection. However, in the presence of polysorbate the recovery

significantly decreased at pH values higher than pH 2 (citrate sys-
tem) and pH 5 (phosphate system), respectively. Thus, the transi-
tion to higher pH values hampered endotoxin recovery. In addition,
the diverging curve progressions (Fig. 3A) indicate an intrinsic ef-
fect of each particular buffer system. Endotoxin recovery kinetics
using different buffer systems such as ethylenediaminetetraacetic
acid (EDTA), citrate and phosphate were studied and are shown in
Fig. 3B. The endotoxin recovery within the described buffer systems
are plotted as a function of time. The system containing EDTA
showed the most rapid activity loss. The loss of activity was slower
under citric conditions and slowest under phosphoric conditions.
After 6 h recovery was below 30% at each condition. As confirmed
before, surfactants are significantly involved in reducing the ac-
tivity of endotoxin in common detection systems. Therefore, the
Fig. 5. Endotoxin masking capacity of citrate-polysorbate formulation. Defined
effects of different surfactants at constant buffer and endotoxin amounts of endotoxin were added to solutions containing 0.05 wt % polysorbate 20
conditions were examined. In Fig. 4, the endotoxin recovery out of and 10 mM sodium citrate and incubated for 7 days at 4  C. Endotoxin spikes were
surfactant solutions (polysorbate 20, polysorbate 80 and octoxynol prepared out of a LPS stock solution containing 10E6 EU/mL. After incubation,
9) in presence and absence of citrate are plotted as a function of endotoxin measurements were performed. The detectable endotoxin concentration
is shown in relation to the spiked endotoxin concentration.
surfactant concentration. In general, all surfactants significantly
reduced endotoxin detectability in the presence of citrate after
seven days of incubation. In absence of citrate, only octoxynol 9 Spiked endotoxin contents of up to 250 EU/mL resulted in no
showed low recoveries at relatively high concentrations after the endotoxin recovery after seven days of incubation. Medium and
incubation period. high-level spikes resulted in very low (<1%) and low endotoxin
Summarizing the results above, nonionic surfactants and com- (<5%) recovery. This illustrates the high masking capacity of com-
plex forming buffer components in combination cause a significant mon formulation matrices and suggests the need for vigilance in
perturbation of endotoxin detection in Limulus-based detection bacterial endotoxin testing under such conditions.
systems. The resulting LER is time-dependent and occurs solely
after a certain period of time. For a final evaluation, the masking 4. Discussion
capacity of such formulation matrices was examined. Endotoxin
was titrated into samples containing a citrate buffer system and When less than 50% of an endotoxin spike into an undiluted
polysorbate. Fig. 5 shows the capacity of such a particular matrix. sample is recovered over time, the detection of endotoxin is
popularly classified as LER. It is supposed that this phenomenon is
caused by endotoxin masking. Therefore it is important to differ-
entiate this phenomenon from test interference. Test interference,
which directly affects the detection system, can be excluded
because it can be corrected by dilution. Yet, in the case of LER,
endotoxin dilutions up to a factor of 10,000 could not overcome
inadequate recovery (Table 1). Furthermore, within very short in-
cubation periods with endotoxin in the sample, the full endotoxin
content could be recovered, which illustrates the functionality of
the detection system. These findings reflect a previous observation,
namely that under certain condition the interference in Limulus-
based detection methods is concentration independent and as-
sumes that the aggregate conformation of LPS is affected and not
the detection system itself [11]. The assumption of alterations in the
aggregate conformation is supported by the time-dependent
appearance of LER. Obviously, the kinetics in Fig. 2 shows a time-
dependent phenomenon, while test interference appears immedi-
ately and is therefore time-independent. This time-dependent
behavior can be illustrated by an alteration of the supramolecular
structure of the amphiphilic LPS. In general, the process of aggre-
gation of amphiphilic molecules can be very variable with respect
to time-scales for structural changes, which can range from sub-
microseconds to several days, weeks and even month [15]. This
also might explain experimental results, in which the masking
phenomenon was not observed, although masking conditions were
present [16].
However, the results also demonstrate that endotoxin recovery is
Fig. 4. Surfactant dependent endotoxin recovery. 100 EU/mL of endotoxin was added affected by the formulation components themselves, even if the
to solutions containing various amounts of polysorbate 20 (A), polysorbate 80 (B) or active pharmaceutical substance, such as a protein, is absent. The
octoxynol 9 (C). Endotoxin recoveries were determined in the presence of 10 mM simultaneous presence of a nonionic surfactant and complex
▫
citrate immediately after preparation ( ) and after incubation for 7 days at room
forming components (chelator) suffices to decrease the detectability
temperature (-). In parallel endotoxin activities were determined in the absence of
▵
citrate, without incubation ( ) and after incubation (:). Endotoxin recovery is shown of endotoxin. The presence of only one of the formulation compo-
as a function of the particular surfactant concentration. nents is not effective in significantly disturbing endotoxin recovery
 J. Reich et al. / Biologicals 44 (2016) 417e422 421

(Table 1). These findings confirm former assumptions of endotoxin The interaction of nonionic surfactants with LPS aggregates is
disaggregation [17,18] and explain the interdependent interaction of favored if the LPS aggregates possess a certain degree of rigidity
surfactant and chelator on endotoxin. Due to the ionic and amphi- (Fig. 4). The latter is controlled, to some extent, by ionic interactions
philic nature of LPS (Fig. 1), complex forming agents disturb the as described above. Under these circumstances the supramolecular
electrostatic interactions and surfactants potentially disturb the structure of LPS is changed into a structure with a lower affinity to
hydrophobic interactions in endotoxin aggregates. Certainly, to the endotoxin sensitive Factor C of the Limulus-based detection
disturb the supramolecular structure of endotoxin a reduced rigidity system resulting in the measurement of a lower activity. Such a
is beneficial. This is controlled by the salt form of LPS, which again structure could be disaggregated LPS due to a molecular excess of
involves the presence of multivalent cations like Ca2þ [19,20]. surfactants. This hypothesis fits well to the observation of Mueller
Consequently, it can be assumed that complex forming agents are in et al., which have shown that disaggregated LPS molecules
competition with negatively charged patches of the endotoxin. (“monomers”) are substantially less active than aggregated LPS in
Therefore the salt bridges between LPS molecules are disturbed, the detection system used [24]. Additionally, Tan et al. proposed a
which should result in a reduced rigidity of endotoxin aggregates, cooperative binding mechanism of LPS to Factor C, which conse-
which in turn facilitates changes in the supramolecular structure. quently requires more than one LPS molecule in close spatial ar-
Thus, the chelating capability of the buffer component is crucial. In rangements [26]. On the other hand, it has been shown that
the presence of EDTA the recovery drops faster than in the presence monomeric LPS shows a higher potency in activating Limulus
of citrate or phosphate based buffer systems (Fig. 3b). Using buffer Amebocyte Lysate assays than aggregated LPS [27]. Under these
components with higher metal complex forming capabilities circumstances, the inadequate detectability might have steric rea-
accelerate masking kinetics. The related metal complex formation sons, in which the activating spots of the LPS (lipid A) are hidden by
constants are directly proportional to the denticity of the ligand surfactant molecules and are not accessible for Factor C.
(rule of thumb [21]). A hexadentate ligand like EDTA forms more In summary, we propose a two-step mechanism of endotoxin
stable metal complexes than a tridentate ligand like citrate. masking. Fig. 6 illustrates the effects of chelating buffer compo-
Furthermore, the equilibrium complex formation ability and the nents and nonionic surfactants on endotoxin. In this mechanism
complex stability of a chelator are pH dependant. At low pH values, the equilibrium LPS structure is shifted to an altered supramolec-
protons are in competition with cations, which hamper formation of ular structure. In its natural state, LPS monomers tend to aggregate
metal complexes [22]. Consequently, masking of endotoxin is due to the hydrophobic interactions between the Lipid A molecules.
affected by the free concentration of protons (pH), which is Additional ionic interactions, formed by divalent cations and
controlled by the buffer system and its particular acid dissociation negatively charged substitutes (e.g.: phosphates) of the LPS in-
constant. This explains the pH dependent endotoxin recovery in crease the rigidity of aggregates. By adding a complex forming
different buffer systems (Fig. 2A). However, complex forming com- agent (I), the salt bridges formed by divalent cations (e.g. Mg2þ)
ponents alone do not result in inadequate endotoxin detection, and LPS are destabilized, leading to a reduced rigidity of the
further amphiphilic components like surfactants are necessary. Due aggregate. The further presence of a surfactant (II) can then change
to the fact that LPS itself is amphiphilic, it tends to aggregate, the initial supramolecular structure by formation of mixed aggre-
because of the low solubility of the hydrophobic fatty acids of Lipid A gates. This structural change leads inevitably to a certain change in
in an aqueous solution [23]. Thus, LPS exhibits certain supramo- detectable activity, as endotoxin activity is dependent on its su-
lecular structures, which influence detectability in Limulus-based pramolecular structure. Due to the ordinary molar excess of com-
detection systems [24]. Structural transitions of amphiphilic sys- plex forming agent and surfactant (micro molar range), compared
tems are affected by a large variety of physical and chemical pa- to endotoxin content (nano molar range), mixed surfactant micelles
rameters. One major principle to control these structures is the head containing monomerized LPS are the most probable resulting su-
group repulsions of self-assembling molecules. They can be affected pramolecular structure.
by co-surfactants, electrolytes, and amphiphilic counter ions [25]. If Within this study, the phenomenon of LER were confirmed in
the masking surfactant (e.g.: polysorbate) intercalates between LPS Limulus-based detection systems and exemplifies a potential
molecules resulting in reduction of head group repulsions, the mechanism of endotoxin masking. The unknown period of endo-
establishment of a new equilibrium is favored and the supramo- toxin presence during a potential event of endotoxin contamination
lecular structure of LPS is altered. in a sample defines the chance of endotoxin recovery. Hence, LER is

Fig. 6. A proposed two-step mechanism of endotoxin masking. Potential equilibration reaction of endotoxin masking, caused by complex forming agents and surfactants, is
schematically illustrated. In a first step, pure endotoxin aggregates are disturbed by chelators increasing the permeability of the aggregate. Then, surfactants interact with endotoxin
by forming mixed aggregates.
422 J. Reich et al. / Biologicals 44 (2016) 417e422

under control of kinetics that governs the formation of mixed [10] Rudbach JA, Milner KC. Reaction of endotoxin and surfactants. III. Effect of
sodium lauryl sulfate on the structure and pyrogenicity of endotoxin. Can J
endotoxin-surfactant aggregates, which make them less prone to
Microbiol 1968 Nov;14:1173e8.
activate the enzymatic reaction of Limulus-based detection sys- [11] Milton DK, Feldman HA, Neuberg DS, Bruckner RJ, Greaves IA. Environmental
tems. In conclusion, capacity experiments have shown that endotoxin measurement: the kinetic limulus assay with resistant-parallel-line
commonly used excipients are capable of masking hazardous estimation. Environ Res 1992 Apr;57:212e30.
[12] Hughes P, Thomas C, Suvarna K, Chi B, Candau-Chacon R, Gomez-Broughton C,
amounts of endotoxin. Currently, it is not known whether masked et al. Low endotoxin recovery: an FDA perspective. BioPharma Asia
endotoxin activates the innate immune system. Bacterial endotoxin 2015;04(4):14e25.
testing should be performed with care, especially in the presence of [13] Williams KL. Endotoxin test concerns of biologics. Am Pharm Rev 2013;16:
4e9.
surfactants and chelators. [14] Daugherty AL, Mrsny RJ. Formulation and delivery issues for monoclonal
antibody therapeutics. Adv Drug Deliv Rev 2006 Aug;58:686e706.
Acknowledgment [15] Gradzielski M. Kinetics of morphological changes in surfactant systems. Curr
Opin Colloid Interface Sci 2003 Nov;8:337e45.
[16] Bolden JS, Claerbout ME, Miner MK, Murphy MA, Smith KR, Warburton RE.
The authors wish to express their gratitude to the participating Evidence against a bacterial endotoxin masking effect in biologic drug prod-
organizations, which enabled a charitable and comprehensive ucts by limulus amebocyte lysate detection. PDA J Pharm Sci Technol 2014 Sep
1;68:472e7.
study between academia and industry. Further, the authors are [17] Tsuchiya M: Possible mechanism of low endotoxin recovery. Am Pharm Rev
grateful to Prof. Ulrich Koszinowski for critical review of this 17.
manuscript. The endotoxin detection kits were a kind gift from [18] Petsch D, Anspach FB. Endotoxin removal from protein solutions. J Biotechnol
2000 Jan 21;76:97e119.
Hyglos GmbH, Bernried, Germany.
[19] Snyder S, Kim D, McIntosh TJ. Lipopolysaccharide bilayer structure: effect of
chemotype, core mutations, divalent cations, and temperature. Biochem
References (Mosc) 1999 Aug;38:10758e67.
[20] Garidel P, Rappolt M, Schromm AB, Howe J, Lohner K, Andr€ a J, et al. Divalent
[1] Seydel U, Schromm AB, Blunck R, Brandenburg K. Chemical structure, mo- cations affect chain mobility and aggregate structure of lipopolysaccharide
lecular conformation, and bioactivity of endotoxins. Chem Immunol 2000;74: from Salmonella minnesota reflected in a decrease of its biological activity.
5e24. Biochim Biophys Acta BBA Biomembr 2005 Sep;1715:122e31.
[2] Fiuza C, Suffredini AF. Human models of innate immunity: local and systemic [21] Holleman AF, Wiberg E, Wiberg N. Holleman-Wiberg: lehrbuch der anorga-
inflammatory responses. J Endotoxin Res 2001;7:385e8. nischen chemie. 1995. 101., verb. und stark erw. Aufl Berlin [u.a.], de Gruyter.
[3] Martich GD, Boujoukos AJ, Suffredini AF. Response of man to endotoxin. [22] Jander G, Blasius E, Str€
ahle J, Schweda E, Rossi R. Jander-Blasius: lehrbuch der
analytischen und pra €parativen anorganischen Chemie. 2006. 16., überarb. Aufl
Immunobiology 1993 Apr;187:403e16.
[4] Petsch D, Deckwer WD, Anspach FB. Proteinase K digestion of proteins im- Stuttgart, Hirzel.
proves detection of bacterial endotoxins by the Limulus amebocyte lysate [23] Schromm AB, Brandenburg K, Rietschel ET, Flad H-D, Carroll SF, Seydel U.
assay: application for endotoxin removal from cationic proteins. Anal Biochem Lipopolysaccharide-binding protein mediates CD14-independent intercala-
1998 May 15;259:42e7. tion of lipopolysaccharide into phospholipid membranes. FEBS Lett 1996
[5] Gnauck A, Lentle RG, Kruger MC. The Limulus Amebocyte Lysate assay may be Dec;399:267e71.
unsuitable for detecting endotoxin in blood of healthy female subjects. [24] Mueller M, Lindner B, Kusumoto S, Fukase K, Schromm AB, Seydel U. Aggre-
J Immunol Methods 2015 Jan;416:146e56. gates are the biologically active units of endotoxin. J Biol Chem 2004 Jun
[6] McCullough KZ. Parenteral drug association: the bacterial endotoxins test: a 18;279:26307e13.
[25] Evans DF, Wennerstro € m H. The colloidal domain: where physics, chemistry,
practical approach. Bethesda, MD, USA; River Grove, IL: PDA; DHI Pub; 2011.
[7] European Pharmacopoeia. In: English subscription to main biology, and technology meet. 2nd ed. New York: Wiley-VCH; 1999.
volume þ supplement 1 þ supplement 2. 8th ed. Stuttgart: Deutscher [26] Tan NS, Ng ML, Yau YH, Chong PK, Ho B, Ding JL. Definition of endotoxin
Apotheker Verlag; 2013. 2013. binding sites in horseshoe crab factor C recombinant sushi proteins and
[8] Hurley JC. Endotoxemia: methods of detection and clinical correlates. Clin neutralization of endotoxin by sushi peptides. FASEB J Off Publ Fed Am Soc
Microbiol Rev 1995 Apr;8:268e92. Exp Biol 2000 Sep;14:1801e13.
[9] Ribi E, Anacker RL, Brown R, Haskins WT, Malmgren B, Milner KC, et al. Re- [27] Takayama K, Din Z, Mukerjee P, Cooke P, Kirkland T. Physicochemical prop-
action of endotoxin and surfactants. I. Physical and biological properties of erties of the lipopolysaccharide unit that activates B lymphocytes. J Biol Chem
endotoxin treated with sodium deoxycholate. J Bacteriol 1966 Nov;92: 1990 Aug;265:14023e9.
1493e509.