Zwitterionic reagents for labeling, cross-linking and improving the performance of chemiluminescent immunoassays

View Article Online / Journal Homepage / Table of Contents for this issue

Dynamic Article Links

Organic &

Biomolecular

Chemistry

Cite this: Org. Biomol. Chem., 2012, 10, 1883

www.rsc.org/obc

PAPER

Zwitterionic reagents for labeling, cross-linking and improving the

performance of chemiluminescent immunoassays†

Anand Natrajan,* David Sharpe and David Wen

Received 27th October 2011, Accepted 6th December 2011

DOI: 10.1039/c2ob06807a

Improving reagent performance in immunoassays both to enhance assay sensitivity and to minimize

interference are ongoing challenges in clinical diagnostics. We describe herein the syntheses of a new

class of hydrophilic reagents containing sulfobetaine zwitterions and their applications. These zwitterionic

reagents are potentially useful for improving the properties of immunoassay reagents. We demonstrate for

the ﬁrst time that zwitterion labeling is a general and viable strategy for reducing the non-speciﬁc binding

of proteins to microparticles and, to improve the aqueous solubility of hydrophobic peptides. We also

describe the synthesis of zwitterionic cross-linking reagents and demonstrate their utility for peptide

conjugation. In automated, chemiluminescent immunoassays, improved assay performance was observed

for a hydrophobic, small analyte (theophylline) using an acridinium ester conjugate with a zwitterionic

sulfobetaine linker compared to a hexa(ethylene)glycol linker. Sandwich assay performance for a large

analyte (thyroid stimulating hormone) was similar for the two acridinium ester labels. These results

indicate that zwitterions are complementary to poly(ethylene)glycol in improving the aqueous solubility

and reducing the non-speciﬁc binding of chemiluminescent acridinium ester conjugates.

ionic polymer p

̲

Introduction

teins exhibit lower non-speciﬁc binding to surfaces⁵and PEG

modiﬁcation can confer beneﬁcial improvements in the pharma-

cokinetic proﬁles of therapeutic proteins.⁶We have shown that

PEG improves the aqueous solubility of chemiluminescent labels

such as acridinium esters and that PEG-modiﬁed acridinium

esters exhibit low non-speciﬁc binding and improved immunoas-

say performance.⁷PEG has also been advocated as a reagent to

mitigate interference in immunoassays from endogenous anti-

bodies.⁸In addition to these applications, PEG has also been

used to devise inert surfaces that resist protein adsorption.⁹

Zwitterions represent a different class of molecules that are

Improving the aqueous solubility of reagents, preventing aggre-

gation, lowering non-speciﬁc binding and minimizing non-

speciﬁc interactions with sample components in serum or whole

blood, are ongoing challenges in clinical diagnostic immuno-

assays. Commercial immunoassays often use hydrophobic che-

miluminescent labels such as luminol or acridinium compounds¹

that have limited aqueous solubility and these labels can exacer-

bate the non-speciﬁc binding of proteins such as antibodies.

Recombinant proteins and polypeptides often exhibit poor

aqueous solubility as well as a tendency to form insoluble aggre-

gates because of misfolding or denaturation thereby limiting

their usefulness in both the pharmaceutical and diagnostics

industry.²Immunoassay reagents can also interact with sample

components in a non-speciﬁc manner giving rise to false posi-

tives in assays that have been attributed to natural, weakly-

binding, polyspeciﬁc antibodies.³Blocking reagents are often

used to alleviate this problem.³

also very hydrophilic and perhaps more so than PEG.¹⁰

ssembled monolayers (SAMs) with zwitterionic functional

Self-

̲

a̲

̲

groups have been shown to be extremely resistant to protein

adsorption.¹¹More recently, a number of studies have reported

that surfaces functionalized with zwitterionic polymers contain-

ing sulfobetaines and carboxybetaines resist protein adsorption

and bioﬁlm formation.¹²A direct comparison of PEG-modiﬁed

and zwitterion-modiﬁed silica surfaces indicated that zwitterions

and PEG are equally effective in resisting protein adsorption.¹³It

has also been postulated that zwitterions would be chemically

more stable than PEG owing to the latter’s propensity for oxi-

dative cleavage.¹¹Zwitterions such as sulfobetaines with qua-

ternary nitrogens are also electrically neutral (non-ionic) over a

wide range of pH.

One approach that has been used to improve the solubility of

peptides and proteins is by labeling with the hydrophilic, non-

Siemens Healthcare Diagnostics, Advanced Technology and Pre-

Development, 333 Coney Street, East Walpole, MA 02032, USA.

E-mail: anand.natrajan@siemens.com; Fax: +1-508-660-4591; Tel:

+1-508-660-4582

†Electronic supplementary information (ESI) available. See DOI:

10.1039/c2ob06807a

In addition to surface modiﬁcation, zwitterionic compounds

such as n

̲

This journal is © The Royal Society of Chemistry 2012

View Article Online

to be useful as additives in assisting refolding of proteins such as

BSA (bovine serum a

lbumin), enzymes and antibodies¹⁴as well

̲

as improving the thermal stability of proteins.¹⁵Tolbert and co-

workers have reported that site-speciﬁc modiﬁcation of two

aggregation-prone polypeptides with betaine (trimethylammo-

nium moiety) signiﬁcantly improved the aqueous solubility of

the two polypeptides and inhibited aggregate formation.¹⁶All

these studies suggest that zwitterionic labeling reagents would

be useful for reducing the non-speciﬁc binding of proteins as

well as alleviating the aqueous solubility of hydrophobic labels

and antigens.

Although a number of PEG-modiﬁed reagents are available

from commercial vendors, analogous zwitterionic reagents have

not been described in the literature. We describe herein a new

class of hydrophilic, labeling and cross-linking reagents contain-

ing sulfobetaine zwitterions that are conveniently synthesized

using commercially available reagents. These zwitterionic

reagents are useful for protein labeling to reduce their non-

speciﬁc binding to microparticles as well as for improving the

aqueous solubility of hydrophobic antigens such as peptides.

Our studies are the ﬁrst to describe zwitterion labeling of pro-

teins as a general strategy for reducing their non-speciﬁc binding

Fig. 1 Structures of zwitterionic reagents.

with 1,3-propane sultone to introduce the sulfobetaine zwitterion

(compound ii). Hydrogenation of the benzyl ester gave com-

pound iii which was converted to the pentaﬂuorophenyl ester 1

and puriﬁed by preparative HPLC. The pentaﬂuorophenyl reac-

tive ester was selected instead of a N-hydroxysuccinimide ester

because it nicely counterbalanced the very polar sulfobetaine

moiety such that the ﬁnal reagent 1 was soluble in polar aprotic

solvents such as DMSO for labeling peptides and proteins.

Moreover the pentaﬂuorophenyl group also made it relatively

easy to purify compound 1 by preparative HPLC with a UV

detector.

The synthesis of compound 2, a thiol-reactive zwitterionic

reagent is described in Figure S2 (supplementary material†). The

primary amine in commercially available N,N-ethylenediamine

was protected as the benzyl carbamate iv followed by N-alky-

lation of the tertiary amine in iv using 1,3-propane sultone to

give compound v. Hydrogenation of compound v gave com-

pound vi which was converted to compound 2 by coupling with

commercially available 4-maleimidobutyric acid followed by

HPLC puriﬁcation.

with

practical

applications

especially in

analytical

immunochemistry.

We also describe the preparation of zwitterionic cross-linking

reagents that are useful for peptide conjugation as well as for the

preparation of hydrophilic chemiluminescent acridinium ester

conjugates of hydrophobic analytes such as theophylline.¹⁷

A

comparison of immunoassay performance of acridinium ester

conjugates containing either a PEG linker or a sulfobetaine zwit-

terion linker in both competitive (theophylline) and sandwich

(thyroid stimulating hormone) assays suggest that zwitterions are

more effective than PEG in improving assay performance for

hydrophobic analytes and, are complementary to PEG in sand-

wich assays.

Results and discussion

Synthesis of zwitterionic reagents

The structures of zwitterionic reagents 1–5 synthesized in the

current study are illustrated in Fig. 1 and the syntheses of the

reagents are illustrated in Figures S1–S3 (supplementary

material†). The zwitterionic, nucleophilic, cross-linker 3 was

described previously and can be easily synthesized in gram

quantities in a three step procedure from commercially available

N,N-bis(3-aminopropyl)methylamine.¹⁷We have observed that

this very polar linker is extremely useful for the synthesis of che-

miluminescent acridinium ester conjugates (tracers for immu-

noassays) of hydrophobic analytes where limited aqueous

solubility is a concern. The syntheses of the compounds in

Fig. 1 were generally accomplished in a straightforward manner

using commercially available reagents as described in the exper-

imental section.

Syntheses of the cross-linking reagents 4 and 5 are illustrated

in Figure S3 (supplementary material†). The homobifunctional

cross-linker 4 was synthesized in one step by reacting compound

3¹⁷with d

̲

puriﬁcation. The NHS ester of this cross-linking reagent is quite

stable because of deactivation by the carbamate linkage but was

observed to be readily reactive towards the α-amino group of

peptides.

The heterobifunctional cross-linker 5 was synthesized using

the commercially available lysine derivative N-α-tert-butoxycar-

bonyl-N-ε-benzyloxycarbonyl-lysine-N-hydroxsuccinimide ester

as the scaffold. Initial reaction of this amino acid derivative with

compound iv introduced the sulfobetaine zwitterion to give com-

pound vii. Hydrogenation of the benzyl carbamate in vii gave

compound viii with the exposed ε-amine which was then con-

densed with 4-malemidobutyric acid to give compound ix.

Removal of the tert-butoxycarbonyl protecting group in ix fol-

Compound 1, an amine-reactive, zwitterionic reagent was syn-

thesized from 4-dimethylaminobutyric acid as described in

Figure S1 (supplementary material†). The carboxylic acid was

ﬁrst converted to the benzyl ester i using a published pro-

cedure.¹⁸The tertiary amine in compound i was then N-alkylated

lowed by reaction of the α-amine of x with d

̲

1884 | Org. Biomol. Chem., 2012, 10, 1883–1895

This journal is © The Royal Society of Chemistry 2012

View Article Online

g

̲

Phe-Gly-Gly-Phe respectively, were isolated in excellent yield

by preparative HPLC.

puriﬁed by preparative HPLC.

All the zwitterionic reagents illustrated in Fig. 1, were

observed to be stable when stored at low temperatures

(< −10 °C). They were also readily soluble in polar aprotic sol-

vents such as dimethyl sulfoxide (DMSO) for labeling. (HPLC

chromatograms of the synthetic intermediates and the ﬁnal

puriﬁed materials are shown in Figures S4–S11, supplementary

material†).

Solubility was estimated by adding small quantities of de-

ionized water to each of the lyophilized, zwitterion-labeled

peptide until a clear solution was obtained. As indicated in

Fig. 2, the labeled peptides 6–8 exhibited signiﬁcant solubility in

water at room temperature and at the indicated concentrations

which gave homogeneous, clear and somewhat soapy solutions.

The calculated solubility¹⁹of penta(phenylalanine) is reported to

be 0.029 g L⁻¹(0.039 mM) in unbuffered water. Zwitterion

labeling of the α-amino group with reagent 1 appears to increase

the solubility > 10-fold. The calculated solubility of less hydro-

phobic tetrapeptide, Phe-Gly-Gly-Phe is 2.8 mM whereas the

zwitterion-labeled peptide 7 was observed to have the same solu-

bility. The calculated solubility of penta(leucine) in unbuffered

water is very low at 0.00016 mM (9.3 × 10⁻⁵g L⁻¹) whereas the

zwitterion labeled peptide 8 was observed to be signiﬁcantly

higher at approximately 0.1 mM.

The facile synthesis of the zwitterionic reagent 1 and its reac-

tivity suggests that this reagent could be of general use for

improving the aqueous solubility of hydrophobic antigens con-

taining reactive amines. Alternatively, compound 2, which is a

thiol-reactive zwitterionic reagent can be used to introduce the

water-soluble sulfobetaine zwitterion for sulfhydryl-containing

antigens.

Peptide and protein labeling with zwitterionic compound 1

The labeling of peptides with the zwitterionic reagent 1 is

described in Fig. 2 and Figures S12–S14 (supplementary

material†). We selected three peptides, penta(phenylalanine), the

tetrapeptide Phe-Gly-Gly-Phe and penta(leucine) as representa-

tive of hydrophobic peptides with limited solubility in water and,

studied the reactivity of reagent 1 towards the α-amino group of

these peptides as well as the impact of zwitterion labeling on

their aqueous solubility. The labeling reactions were conducted

in a mixture of an organic solvent (DMSO) and sodium carbon-

ate buffer, pH = 9, using a slight excess (approximately 2

equivalents) of zwitterion label 1. Under these typical conditions

used for peptide and protein labeling, hydrolysis of the pen-

taﬂuorophenyl ester in zwitterion reagent 1 competes with reac-

tion with the α-amino group of the peptides. As can be noted

from Figures S12 and S13, the α-amino group in the peptides

penta(phenylalanine) and Phe-Gly-Gly-Phe reacted readily with

compound 1 affording excellent conversion. Conversion was

lower (∼60%, Figure S14), for the extremely hydrophobic

peptide penta(leucine) presumably because of poorer solubility,

accounting for the lower isolated yield (43%) of the zwitterion-

labeled peptide 8. On the other hand, the other two zwitterion-

labeled peptides 6 and 7, derived from penta(phenylalanine) and

To assess the reactivity of reagent 1 towards proteins, we

selected b

̲

by MALDI-TOF mass spectroscopy for measurement of label

incorporation. Following light labeling using 5 equivalents input

of the acridinium ester 9-[[4-[[(2,5-dioxo-1-pyrrolidinyl)oxy]car-

bonyl]-2,6-dimethylphenoxy]carbonyl]-10-(3-sulfopropyl)-

acridinium (abbreviated as NSP-DMAE-NHS,⁷(Table 1), the

labeled protein containing 1.4 labels, was further reacted with

compound 1. Using an input of 10 and 20 equivalents of com-

pound 1, mass spectral analysis of the labeled protein indicated

Fig. 2 Hydrophobic peptide labeling with the zwitterionic reagent 1. Reagents: (a) dimethyl sulfoxide, 0.1 M sodium carbonate, pH 9.

This journal is © The Royal Society of Chemistry 2012 Org. Biomol. Chem., 2012, 10, 1883–1895 | 1885

View Article Online

Table 1 Non-speciﬁc binding of zwitterion-labeled proteins

Fractional non-speciﬁc bindings for 0.070 mg of three solid phases of 1 pmol zwitterion-modiﬁed proteins labeled with NSP-DMAE

NSP-DMAE-AVD

Zwitterion-

NSP-DMAE-BGG

Zwitterion-

NSP-DMAE-FBN

Protein

FNSB

Reduction

(fold)

FNSB

Reduction

(fold)

Zwitterion-

modiﬁed

FNSB

Reduction

(fold)

Unmodiﬁed

modiﬁed

FNSB (%)

Unmodiﬁed

modiﬁed

FNSB (%)

Unmodiﬁed

FNSB (%)

Solid phase

FNSB (%)

PMP

0.13

0.017

7.6

0.012

0.0068

1.8

0.0094

0.0071

1.3

derivatized

with sheep

IgG

Dynalbeads

M-280

streptavidin

Dynalbeads

M-270

0.016

6.3

0.0083

0.052

1.9

0.013

0.18

0.0081

0.039

1.6

4.6

0.0062

0.037

0.0063

0.012

1.0

3.1

121

streptavidin

Abbreviations used: PMP = p

̲

in; BGG = b

̲

the incorporation of 5 and 8 zwitterion labels respectively.

Similar results were noted using a mixture of bovine gamma glo-

While numerous studies have delineated the properties of

zwitterion-modiﬁed surfaces,^11–13there are no studies that have

examined the impact of zwitterion labeling of proteins on their

non-speciﬁc binding to surfaces. Moreover, while studies on

protein adsorption to well deﬁned surfaces offer important theor-

etical insights, nevertheless, in the practice of clinical diagnostics

immunoassays (using chemiluminescent or ﬂuorescent labels

and magnetic microparticles), such simple situations are rarely

encountered. Instead, microparticles are commonly used as the

scaffold for the immobilization of a variety of binding molecules

such as antibodies with different isoelectric points (pIs) or

various antigens. Blocking proteins and surfactants are also

usually added leading to a chemically heterogeneous, particle

surface. Adding to the complexity is the immunoassay sample

which is usually serum or whole blood.

̲

bulins (BGG) as described in the experimental section. The reac-

tivity of the pentaﬂuorophenyl ester in reagent 1 towards

proteins thus appears similar to the reactivity of N-hydroxysucci-

nimide (NHS) ester derivatives i.e. > 40% label incorporation.

Zwitterion-modiﬁed surfaces resist protein adsorption¹²and

recently, Schlenoff and co-workers,¹³from a study of PEG and

zwitterion-functionalized silica surfaces, have proposed a model

wherein the main driving force for adsorption is largely entropic

in nature. According to this model, protein adsorption to surfaces

occurs due to the release of counter ions from both the protein

and the surface during the adsorption process. The net free

energy change is negative for such a process. For surfaces that

are non-ionic, such as those modiﬁed by PEG or zwitterions,

this mechanism is absent and consequently, the surfaces are

resistant to protein adsorption. Regardless of the mechanism,

based on an earlier empirical study, Whitesides and co-workers¹¹

outlined some general principles for designing a protein-resistant

surface and they include (a) hydrophilic nature to minimize

hydrophobic interactions, (b) overall electrical neutrality to avoid

charge interactions and, (c) ability to accept but not donate

hydrogen bonds. We have noted that these same principles are

also useful for the modiﬁcation of hydrophobic chemilumines-

cent labels such as acridinium esters to lower their non-speciﬁc

binding and to improve their assay performance.⁷

In the current study, we wanted to determine whether zwitter-

ion-labeling of proteins was a general strategy that could be

employed for reducing their non-speciﬁc binding to magnetic

microparticles that are commonly used in immunoassays. High

non-speciﬁc binding is often encountered in immunoassays and

is a limiting factor in improving assay sensitivity. We selected

three proteins, egg white a

̲

ovine g

̲

(BGG) and fibrinogen (FBN) to study non-speciﬁc binding to

̲

microparticles that are commonly used in immunoassays. AVD is

a basic protein²⁰and the avidin–biotin interaction is commonly

used in immunoassays. BGG is representative of polyclonal

1886 | Org. Biomol. Chem., 2012, 10, 1883–1895

This journal is © The Royal Society of Chemistry 2012

View Article Online

antibodies that are commonly used in immunoassays. Finally,

FBN is representative of a large protein that has been used to

evaluate the properties of surfaces towards protein adsorption.¹¹

two orders of magnitude to the M-270 microparticles and, by a

factor of two for the relatively more hydrophobic M-280 par-

ticles. A similar trend was observed for BGG although the mag-

nitude of reduction of non-speciﬁc binding was attenuated for

this protein. For example on PMP and M-280 particles, an

approximately 2-fold reduction in non-speciﬁc binding was

observed whereas on M-270 particles, a 4.6-fold reduction in

non-speciﬁc was noted. On the other hand the larger protein

FBN showed only a slight attenuation of its non-speciﬁc binding

to PMP and M-270 particles with zwitterion modiﬁcation but the

trend was similar to that observed for AVD and BGG.

The results outlined in Table 1 have broad implications. These

results suggest that labeling of hydrophobic and/or basic proteins

with zwitterionic reagents 1 (or 2) is a viable and general strat-

egy for reducing their non-speciﬁc binding to surfaces regardless

of the nature of the surface.

For particles, we selected p

̲

well as two magnetic latex particles (MLPs) coated with strepta-

̲

vidin that are available from a commercial vendor (Invitrogen).

The PMPs used in the current evaluation were, 1–10 micron-

sized iron(^III) oxide particles, coated with an anti-TSH antibody

(TSH = t

̲

particle surface using commonly used glutaraldehyde coupling

chemistry. The MLPs used for the current evaluation were

2.8 micron, Dynal^®M-280-streptavidin particles with bound bio-

tinylated, polyclonal, goat anti-PTH antibody (abbreviated as

“M-280” in Table 1) and Dynal^®M-270-streptavidin particles

with bound biotinylated, monoclonal, mouse anti-cTnI antibody

(abbreviated as “M-270” in Table 1). These two MLPs had

immobilized streptavidin on their surfaces which was then used

to further immobilize, biotin-labeled antibodies capable of

binding the analytes PTH (p

̲

Peptide cross-linking using zwitterionic cross-linking reagents 4

and 5

(cardiac troponin I) respectively. The intrinsic, surface character-

̲

istics of the three types of particles are quite different because of

different functional groups. PMPs contain amines on their sur-

faces whereas M-270 particles have carboxylate surfaces. The

M-280 particles have a more hydrophobic surface with a tosy-

late-activated polystyrene coating (Experimental section). The

surfaces of all three particles were also immobilized with differ-

ent proteins thus creating very heterogeneous surfaces.

Cross-linking agents are commonly used for the preparation of

conjugates and a variety of these reagents are available from

commercial vendors. Typical cross-linking reagents contain two

reactive groups such as electrophilic, nucleophilic or photoreac-

tive groups that can either be the same (homobifunctional) or

different (heterobifunctional). Most cross-linking reagents tend

to be hydrophobic although some hydrophilic, PEG-modiﬁed

reagents with oligo(ethylene)glycol linkers to alleviate aqueous

solubility, are also available from commercial vendors. The two

zwitterionic cross-linking reagents 4 and 5 (Fig. 1) however are

much more polar with excellent aqueous solubility because of

the strongly hydrophilic nature of the sulfobetaine zwitterion.

We envision that these reagents would be especially useful for

cross-linking hydrophobic antigens while at the same time

improving their aqueous solubility.

The cross-linking of representative, hydrophobic synthetic

peptides using the zwitterionic cross-linking reagents 4 and 5 is

illustrated in Fig. 3 and 4. The cross-linking of the α-amino

groups of the two peptides penta(phenylalanine) and the tetra-

peptide Phe-Gly-Gly-Phe using reagent 4 containing two,

amine-reactive, NHS ester groups is described in Fig. 3. Initial

reaction of penta(phenylalanine) with a slight excess of 4

resulted in clean conversion to the labeled peptide intermediate

xi which could be puriﬁed by preparative HPLC without com-

promising the second, reactive NHS ester in xi. Subsequent reac-

tion with the second peptide Phe-Gly-Gly-Phe resulted in clean

formation of the nonapeptide 9 with the zwitterionic reagent

inserted between the two peptides (Figure S15, supplementary

material†).

The three proteins AVD, BGG and FBN were initially lightly

labeled (5 equivalents input) with the chemiluminescent acridi-

nium ester NSP-DMAE-NHS,⁷(Table 1). Chemiluminescence

from the acridinium ester label was used to measure non-speciﬁc

binding of the labeled proteins. A portion of each acridinium

ester labeled conjugate was further labeled with 25 equivalents

input of the zwitterionic compound 1 as described in the exper-

imental section. The labeled conjugates, with and without zwit-

terion modiﬁcation, were used for measurement of non-speciﬁc

binding to the microparticles by recording the chemilumines-

cence from the acridinium ester label. The two types of particles

(PMPs and MLPs) were mixed with solutions of the acridinium

ester and zwitterion-labeled conjugates and were then magneti-

cally separated, washed twice with water and then the chemilu-

minescence associated with the particles was measured. The

results of these measurements are tabulated in Table 1 and

details are described in the experimental section. The ratio of

this chemiluminescence value in comparison to the total chemi-

luminescence input is referred to f

̲

(FNSB) and, in this case is a reﬂection of the resistance of the

protein towards non-speciﬁc adsorption to the microparticles.

FNSB is deﬁned similarly in assays for analytes and is simply

the ratio of background signal to total chemiluminescence input.

For small analytes, background signal is measured at extremely

high analyte concentration whereas for sandwich assays, chemi-

luminescence in the absence of analyte is the background signal.

(Details of these measurements can be found in the experimental

section.)

In a similar vein, cross-linking of the peptides penta

(phenylalanine) and the sulfhydryl-containing dipeptide Cys-

Gly using the zwitterionic reagent 5 was equally facile as

described in Fig. 4. Initial reaction of penta(phenylalanine) with

5 gave intermediate xii with its maleimide group intact and sub-

sequent reaction (without HPLC puriﬁcation) with the sulfyhy-

dryl-containing dipeptide resulted in quantitative conversion to

the heptapeptide 10 containing the sulfobetaine zwitterion

(Figure S16, supplementary material†).

As indicated in Table 1, zwitterion-labeling of the basic

protein AVD and BGG had a signiﬁcant impact on their non-

speciﬁc binding to the microparticles. For AVD, labeling with

reagent 1 reduced its FNSB by an order of magnitude to PMP,

This journal is © The Royal Society of Chemistry 2012

Org. Biomol. Chem., 2012, 10, 1883–1895 | 1887

View Article Online

Fig. 3 Synthetic scheme of hydrophobic peptide cross-linking using zwitterionic, homobifunctional cross-linking reagent 4. Reagents: (a, b)

dimethyl sulfoxide, diisopropylethylamine.

Fig. 4 Synthetic scheme for hydrophobic peptide cross-linking using zwitterionic, heterobifunctional cross-linking reagent 5. Reagents: (a) dimethyl

formamide, triethylamine; (b) dimethyl formamide.

Immunoassay performance

used to enhance the sensitivity of automated, chemiluminescent,

immunoassays.⁷In the current study we compared the effect of

two hydrophilic linkers containing either PEG or a sulfobetaine

zwitterion on immunoassay performance of acridinium ester

labels in both competitive (theophylline) and sandwich (TSH)

We reported previously that PEG-modiﬁed, acridinium ester

labels containing alkoxy groups in the acridinium ring exhibit

increased output and that these high light output labels can be

1888 | Org. Biomol. Chem., 2012, 10, 1883–1895

This journal is © The Royal Society of Chemistry 2012

View Article Online

prepared using the zwitterionic, nucleophilic cross-linker 3 in

Fig. 1. Both conjugates 11 and 12 exhibited similar chemilumi-

nescence proﬁles.¹⁷

Assay performance of conjugates 11 and 12 was compared on

Siemens’ Healthcare Diagnostics’ ADVIA Centaur^®system, an

automated immunochemistry analyzer. In the assay, acridinium

ester–theophylline conjugates are used to measure the concen-

tration of theophylline in a sample through titration of the

remaining unoccupied theophylline binding sites of anti-theo-

phylline antibody immobilized on paramagnetic particles. As

indicated in Fig. 5 (panel A), and Table S1 (supplementary

material†) in the assay, the measured chemiluminescent signal is

inversely correlated with the concentration of theophylline. In

the assay comparing the two conjugates 11 and 12, the latter con-

jugate with the zwitterion-containing linker exhibited a steeper

slope at low dose with equivalent precision, and lower FNSB at

high dose. The observed FNSB of 0.39% for conjugate 12 was

observed to be lower than that of conjugate 11 whose FNSB was

measured to be 0.61%. Thus, by replacing conjugate 11 with 12

in the automated assay, the limit of functional sensitivity

decreased from 4 μM to 2 μM reﬂecting a twofold improvement.

Fig. 5 Structures of chemiluminescent acridinium ester–theophylline

conjugates 11 and 12 (panel B) and dose–response curve (panel A) in

the automated theophylline assay using these conjugates. Conjugate 12

with zwitterion linker exhibited lower non-speciﬁc binding in the assay

and improved assay sensitivity. T is the total chemiluminescence input;

B₀is the chemiluminescence in the absence of theophylline and B is the

chemiluminescence in the presence of theophylline. Arrows in the low-

dose region of the graph indicate the limit of detection which was

observed to be 2 μM for conjugate 12 and 4 μM for conjugate 11 reﬂect-

ing a two-fold improvement in assay sensitivity. Dotted lines indicate

FNSB of conjugates 11 (black line) and 12 (blue line). Standard devi-

ations for assay data were calculated in Microsoft Excel using the built

in standard deviation function (STDEV) for ﬁve replicates. Microsoft

Excel uses the following function to calculate standard deviation

0

1

sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ

P

2

ðx ꢀ xˉÞ

ðn ꢀ 1Þ

@

A

:

immunoassays. The purpose of this study was to determine

whether the nature of the hydrophilic linker (PEG or zwitterion)

would affect assay performance.

Theophylline assay

Fig. 6 Structures of chemiluminescent acridinium ester labels 13 and

14 (panel B) and dose–response curve (panel A) in the automated TSH

assay using these labels. Both labels had identical assay performance

with functional sensitivity of 0.02 mIU L⁻¹indicated by the arrows at

the low dose end. Dotted lines indicate FNSB of label 13 (black line)

and label 14 (blue line). Standard deviations for assay data were calcu-

lated in Microsoft Excel using the built in standard deviation function

(STDEV) for ﬁve replicates. Microsoft Excel uses the following function

to calculate standard deviation

Theophylline, is representative of a hydrophobic analyte that is

commonly measured by immunoassay. In the assay, we com-

pared the immunoassay performance of chemiluminescent acri-

dinium ester–theophylline conjugates containing either a PEG

linker (compound 11, Fig. 5, panel B) or containing a sulfo-

betaine zwitterion (compound 12, Fig. 5, Panel B). As can be

noted, the two conjugates are structurally very similar, the only

difference being the structure of the linker used to conjugate the

acridinium ester to the analyte theophylline.

0

1

sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ

P

2

ðx ꢀ xˉÞ

ðn ꢀ 1Þ

@

A

:

The syntheses of the two acridinium ester–theophylline conju-

gates have been reported previously.^7,17Conjugate 12 was

This journal is © The Royal Society of Chemistry 2012

Org. Biomol. Chem., 2012, 10, 1883–1895 | 1889

View Article Online

Both conjugates 11 and 12 also exhibited identical binding

afﬁnity of 47% and 46% (B₀/T%) respectively for the antibody

immobilized on the solid phase. Our results clearly demonstrate

that, in this instance, the sulfobetaine cross-linker is more effec-

tive than PEG in improving assay performance and could be

useful for other hydrophobic analytes as well.

speciﬁc binding and improving immunoassay performance of

hydrophobic analytes. Our studies on protein and peptide label-

ing also indicate that zwitterion labels are useful for decreasing

the non-speciﬁc binding of hydrophobic and basic proteins as

well as alleviating the aqueous solubility of hydrophobic pep-

tides. Although not a focus of our research, we believe, the rela-

tively simple to assemble, zwitterionic reagents that we have

described should be useful for other applications as well such as

the modiﬁcation of the surfaces of nanomaterials to make them

more hydrophilic.

TSH assay

TSH is a clinically important marker for thyroid function and is

commonly measured by immunoassay. In the current assay, TSH

was measured in a sandwich format using two different anti-

bodies directed against different epitopes on the TSH molecule,

where the capture antibody resided on the solid phase and the

signaling antibody was labeled with acridinium ester. Under

identical conditions, we compared the performance of the acridi-

nium ester label compound 13 with that of the label compound

14 (Fig. 6, panel B), labeled to the same extent (4 labels per anti-

body) on the same signaling antibody, a murine monoclonal anti-

body with an acidic pI = 5.6. The syntheses of compounds 13

and 14 were described previously.^7,17Light output from the two

conjugates was similar.¹⁷The observed dose responses in the

TSH assay are shown in Fig. 6 (panel A) and Table S2 (sup-

plementary material†). In the assay, FNSB (0.012%) was

observed to be the same for both conjugates of compounds 13

and 14 indicating that the sulfobetaine linker in compound 14

had a similar effect on non-speciﬁc binding to paramagnetic par-

ticles as the PEG linker in compound 13. As a consequence of

both equivalent assay precision for both labels over the whole

standard curve, the combination of equivalent light output and

equivalent FNSB, equivalent functional sensitivity in the TSH

assay of 0.020 mIU L⁻¹TSH was observed for conjugates of the

two labels 13 and 14 (1 IU TSH = 1.58 × 10⁻⁸mol). Thus, in

the sandwich assay, chemiluminescent acridinium esters contain-

ing either PEG or a sulfobetaine linker were equally effective in

reducing the non-speciﬁc binding of the labeled antibody and

consequently, similar assay performance was observed. Whether

these observations would translate to other sandwich assays is

difﬁcult to predict, but our earlier observations on the non-

speciﬁc binding of proteins indicate that antibodies that are more

hydrophobic or that are basic would beneﬁt more by labeling

with a hydrophilic, zwitterionic chemiluminescent label.

Experimental

General

Chemicals were purchased from Sigma-Aldrich (Milwaukee,

Wisconsin, USA) unless indicated otherwise. All ﬁnal com-

pounds were analyzed and puriﬁed by HPLC using a Beckman-

Coulter HPLC system. HPTLC of synthetic compounds was per-

formed using silica gel 60 CN F₂₅₄S plates from EMD Chemi-

cals Inc. An iodine chamber was used for visualization of spots.

MALDI-TOF (M

̲

T̲

̲

Voyager DE™ Biospectrometry™ Workstation from Perkin-

Elmer. This is a bench top instrument operating in the linear

mode with a 1.2 meter ion path length, ﬂight tube. Spectra were

acquired in positive ion mode. For small molecules, α-cyano-4-

hydroxycinnamic acid was used as the matrix and spectra were

acquired with an accelerating voltage of 20 000 volts and a delay

time of 100 ns. For protein conjugates, sinapinic acid was used

as the matrix and spectra were acquired with an accelerating

voltage of 25 000 volts and a delay time of 85 ns. For HRMS

(H

̲

HPLC-grade methanol and analyzed by direct-ﬂow injection

(injection volume = 5 μL) ElectroSpray Ionization (ESI) on a

Waters Qtof API US instrument in the positive ion mode. Opti-

mized conditions were as follows: Capillary = 3000 kV, Cone =

35, Source T = 120 °C, Desolvation T = 350 °C. NMR spectra

were recorded on a Varian 500 MHz spectrometer.

1. Synthesis of zwitterionic reagent 1 (Figures S1, S4 and S5,

supplementary material†)

Compound i. Compound i was synthesized from 4-dimethyla-

minobutyric acid using a literature procedure.¹⁸

Compound ii. A solution of compound i (0.2 g, 0.905 mmol)

in anhydrous DMF (5 mL) was treated with 1,3-propane sultone

(0.165 g, 1.5 equivalents) and 2,6-di-tert-butylpyridine

(0.24 mL, 1.2 equivalents). The reaction was heated at 150 °C

under a nitrogen atmosphere. After 1 h, HPLC analysis was per-

formed using a Phenomenex, 10 micron, C₁₈, 3.9 × 300 mm

column and a 40 min gradient of 10 → 40% MeCN/water (each

with 0.05% triﬂuoroacetic acid, TFA) at a ﬂow rate of 1.0 mL

min⁻¹and UV detection at 260 and 220 nm. Product was

observed eluting at 14 min with a minor amount of starting

material at 13 min. The reaction was cooled to room temperature

and the solvent was removed under reduced pressure. The

residue was partitioned between ethyl acetate (50 mL) and water

(50 mL). The aqueous solution was separated and washed with

ethyl acetate (2 × 50 mL). The aqueous solution was treated with

Conclusions

We have described the syntheses and applications of new label-

ing and cross-linking reagents containing hydrophilic zwitter-

ions. These reagents are easily assembled and are useful for

improving the aqueous solubility of hydrophobic antigens as

well as for their cross-linking. We had previously reported that

PEG is useful for reducing the non-speciﬁc binding of chemi-

luminescent acridinium ester labels and, that linkers derived

from PEG are useful for the preparation of acridinium ester con-

jugates of hydrophobic analytes such as folate and theophylline.⁷

Our current results comparing the immunoassay performance of

acridinium ester–theophylline conjugates suggest that linkers

containing sulfobetaine zwitterions would be more effective than

PEG in increasing the aqueous solubility, lowering the non-

1890 | Org. Biomol. Chem., 2012, 10, 1883–1895

This journal is © The Royal Society of Chemistry 2012

View Article Online

5 drops of ammonium hydroxide and further extracted with ethyl

acetate (2 × 25 mL). The aqueous solution was then concentrated

under reduced pressure to afford a sticky gum. Yield = 0.35 g

(quantitative); HPTLC (3 : 7, MeCN : water with 0.05% TFA)

white solid was dried under high vacuum to give 1.31 g (85%)

of desired product. HPTLC (3 : 7, MeCN : water with 0.05%

TFA) R_f0.59; H-NMR (500 MHz, CD₃OD) δ 2.20 (m, 2H),

2.85 (t, 2H, J = 6.7), 3.14 (s, 6H), 3.45 (t, 2H, J = 6.6),

3.52–3.62 (m, 4H), 5.11 (s, 2H), 7.27–7.39 (m, 5H); MALDI-

TOF MS m/z 344.9 (M + H)⁺; HRMS m/z 345.1487 (M + H)⁺

(345.1484 calculated).

1

R_f0.55; H-NMR (500 MHz, CF₃COOD) δ 2.19 (m, 2H), 2.43

(m, 2H), 2.64 (t, 2H, J = 6.7), 3.07 (s, 6H), 3.34 (m, 4H), 3.59

(m, 2H), 5.23 (s, 2H), 7.33 (s, 5 H); MALDI-TOF MS m/z 344.3

(M + H)⁺; HRMS m/z 344.1533 (M + H)⁺(344.1532 calculated).

Compound 2. Compound v (580 mg, 1.7 mmol) was dissolved

in methanol/water (95/5, 40 mL) and 10% palladium on carbon

(58 mg) was added. The reaction was hydrogenated for 4 h at

room temperature using a Parr shaker. The reaction mixture was

then ﬁltered and the ﬁltrate was concentrated to dryness to give a

white solid as the desired product vi which was used as such for

the next reaction. Yield = 0.35 g (quantitative). MALDI-TOF

MS m/z 211.0 (M + H)⁺(211.1 calculated).

Compound 1. A solution of compound ii (0.42 g, 1.22 mmol)

in methanol (40 mL) was treated with 0.25 g of 10% palladium

on carbon. The suspension was hydrogenated in a Parr shaker at

40 psi for 2 h at room temperature. The reaction was then ﬁltered

and the methanol solution was concentrated under reduced

pressure to afford a white solid, compound iii which, was used

as such for the next reaction. Yield = 0.234 g (quantitative);

MALDI-TOF MS m/z 254.3 (M + H)⁺(254.1 calculated).

To a solution of compound vi (21.0 mg, 0.1 mmol) and 4-malei-

midobutyric acid (18.3 mg, 0.1 mmol) in anhydrous dimethyl

sulfoxide (1 mL) was added diisopropylethylamine (0.035 mL,

A solution of compound iii (0.114 g, 0.45 mmol) in a

1 : 1 mixture of 0.1 N HCl/MeCN (6 mL) was treated with pen-

taﬂuorophenol (0.125 g, 1.5 equivalents) and 1-ethyl-(dimethyl-

aminopropyl)carbodiimide hydrochloride (EDC·HCl, 0.230 g,

2.5 equivalents). The reaction was stirred at room temperature.

After 1 h, HPLC analysis was performed using a Phenomenex,

C₁₈, 10 micron, 3.9 × 300 mm column and a 40 min gradient of

10 → 60% MeCN/water (with 0.05% TFA) at a ﬂow rate of

1.0 mL min⁻¹and UV detection at 260 and 220 nm. Product

was observed eluting at 18 min. The crude reaction mixture was

puriﬁed by HPLC using a YMC, 10 micron, C₁₈, 30 × 250 mm

column and a 40 min gradient of 10 → 60% MeCN/water (with

0.05% TFA) at a ﬂow rate of 20 mL min⁻¹. The HPLC fractions

containing product were frozen at −80 °C and lyophilized to

dryness to afford a gummy solid. Yield = 78 mg (41%);

¹H-NMR (500 MHz, CF₃COOD) δ 2.38 (m, 2H), 2.59 (m, 2H),

2.98 (t, 2H, J = 6.7), 3.25 (s, 6H), 3.45 (t, 2H, J = 6.9), 3.58 (m,

2H), 3.77 (m, 2H); ¹⁹F-NMR (470 MHz, CF₃COOD) δ −156.6

(d, 1F, J = 17.6), −159.8 (t, 2F, J = 19.8), −165.0 (t, 2F, J =

18.3); MALDI-TOF MS m/z 420.0 (M + H)⁺; HRMS m/z

442.0720 (M + Na)⁺(442.0724 calculated).

0.2 mmol) and b

̲

phonium hexaﬂuorophosphate (BOP, 66.2 mg, 0.15 mmol). The

reaction was stirred at room temperature for 1 h. The crude reac-

tion mixture was puriﬁed by HPLC using a YMC, 10 micron,

C₁₈, 30 × 250 mm column and a 40 min gradient of 0 → 40%

MeCN/water (with 0.05% TFA) at a ﬂow rate of 20 mL min⁻¹

.

The HPLC fractions containing product were frozen at −80 °C

and lyophilized to dryness. Yield = 43 mg (quantitative);

HPTLC (3 : 7, MeCN : water with 0.05% TFA) R_f0.49;

¹H-NMR (500 MHz, CF₃COOD) δ 2.00 (m, 2H), 2.43 (t, 2H,

J = 7.5), 2.49–2.57 (m, 2H), 3.24 (s, 6H), 3.39 (t, 2H, J = 7.0),

3.61 (m, 2H), 3.66 (t, 2H, J = 6.6), 3.72 (m, 2H), 3.89 (m, 2H),

6.87 (s, 2H); MALDI-TOF MS m/z 376.7 (M + H)⁺; HRMS m/z

376.1544 (M + H)⁺(376.1542 calculated).

3. Synthesis of compound 4 (Figures S3 and S8, supplemen-

tary material†). A solution of compound 3¹⁷(25 mg, 56 μmol)

in DMSO (1 mL) was treated with diisopropylethylamine

(0.039 mL, 4 equivalents) and this solution was added drop wise

to a solution of d

̲

2. Synthesis of zwitterionic reagent 2 (Figures S2, S6 and S7,

supplementary material†)

0.37 mmol) in DMSO (2 mL). The reaction was stirred at room

temperature. After 30 min, a portion of the reaction mixture was

analyzed by HPLC using a Phenomenex, 10 micron, C₁₈,

3.9 × 300 mm column and a 30 min gradient of 0 → 40%

MeCN/water (with 0.05% TFA) at a ﬂow rate of 1.0 mL min⁻¹

and UV detection at 220 nm. Product was observed eluting at

10 min. The crude reaction mixture was puriﬁed by HPLC using

a YMC, 10 micron C₁₈, 30 × 250 mm column and a 40 min gra-

dient of 0 → 40% MeCN/water (with 0.05% TFA) above at a

ﬂow rate of 20 mL min⁻¹. The HPLC fractions containing

product were frozen at −80 °C and lyophilized to dryness. Yield

= 15 mg (50%); HPTLC (3 : 7, MeCN : water with 0.05% TFA)

Compound v. To a stirred solution of N-(benzyloxycarbony-

loxy)succinimide (6.30 g, 24.79 mmol, Aldrich) in chloroform

(25 ml) under a nitrogen atmosphere and at 0 °C was added drop

wise N,N-dimethylethylenediamine (2.00 g, 21.55 mmol) dis-

solved in chloroform (15 mL). The reaction mixture was stirred

at 0 °C for 1 h and at room temperature for 1 h. The resulting

reaction mixture was diluted with chloroform (40 mL) and

washed with saturated aqueous sodium bicarbonate (3 × 20 mL).

The organic layer was separated and dried over anhydrous

sodium sulfate. The ﬁltrate was concentrated to give 5.16 g

(quantitative) of viscous oil as the desired product iv which was

used as such for the next reaction. MALDI-TOF MS m/z 223.5

(M + H)⁺(223.1 calculated).

Compound iv (1.00 g, 4.5 mmol) was dissolved in anhydrous

ethyl acetate (5 mL) in a sealed tube and 1,3-propane sultone

(1.10 g, 9.0 mmol) was added under a nitrogen atmosphere. The

sealed tube was heated to 90 °C for 16 h and then cooled to

room temperature. The precipitated product was ﬁltered and

washed with anhydrous ethyl acetate (3 × 20 mL). The resulting

1

R_f0.73; H-NMR (500 MHz, CF₃COOD) δ 2.33 (m, 4H), 2.55

(m, 2H), 3.02 (s, 8H), 3.16 (s, 3H), 3.41 (br t, 2H), 3.52 (m,

6H), 3.62 (m, 2H), 3.69 (m, 2H); MALDI-TOF MS m/z 549.8

(M + H)⁺; HRMS m/z 572.1639 (M + Na)⁺(572.1638 calculated).

4. Synthesis of compound 5 (Figures S3, S9–S11, supplemen-

tary material†)

Compound vii. A solution of compound iv (105 mg,

0.5 mmol) in anhydrous DMSO (2 mL) was treated with

This journal is © The Royal Society of Chemistry 2012

Org. Biomol. Chem., 2012, 10, 1883–1895 | 1891

View Article Online

triethylamine (0.070 mL, 0.5 mmol) and N-α-t-BOC-N-ε-CBZ-

lysine-N-hydroxsuccinimide ester (239 mg, 0.5 mmol, Aldrich).

The suspension was stirred at room temperature for 24 h and

then analyzed by HPLC using a YMC, 3 micron, C₁₈, 4.0 ×

50 mm column and a 10 min gradient of 10 → 90% MeCN/

water (with 0.05% TFA) at a ﬂow rate of 1.0 mL min⁻¹and UV

detection at 260 nm. Product was observed eluting at 6.2 min.

The product was puriﬁed by preparative HPLC using a YMC,

10 micron, C₁₈, 30 × 250 mm column and a 30 min gradient of

10 → 70% MeCN/water (with 0.05% TFA) at a ﬂow rate of

20 mL min⁻¹. The product fraction was collected, frozen at

−80 °C and lyophilized. The product, compound vii, was

obtained as a white powder. Yield = 225 mg (79%); HPTLC

(3 : 7, MeCN : water with 0.05% TFA) R_f0.31; ¹H-NMR

(500 MHz, CF₃COOD) δ 1.59 (s, 9H), 1.61 (m, 2H), 2.00 (m,

2H), 2.42 (m, 2H), 3.20–3.27 (m, 9H), 3.37 (m, 2H), 3.59–3.79

(m, 6H), 4.26 (m, 1H), 4.37 (m, 1H), 5.19 (s, 2H), 7.30 (m, 5H);

MALDI-TOF MS m/z 572.6 (M + H)⁺; HRMS m/z 573.2955 (M

+ H)⁺(573.2958 calculated).

pressure to afford a white sticky solid which was used as such in

the next reaction. Yield = 45 mg (quantitative); MALDI-TOF

MS m/z 504.4 (M + H)⁺(504.3 calculated).

A solution of crude compound x (24 mg, 0.0386 mmol) and dis-

uccinimidyl glutarate (DSG, 50 mg, 4 equivalents, Thermo) in

anhydrous DMSO (2 mL) was treated with triethylamine (6 μL,

1.1 equivalents). The reaction was stirred at room temperature.

After 15 min, HPLC analysis was performed as described pre-

viously. Product was observed eluting at 24 min. The product

was puriﬁed by preparative HPLC as described previously. The

HPLC fractions containing product were frozen at −80 °C and

lyophilized to dryness. Yield = 8 mg (30%); HPTLC (3 : 7,

1

MeCN : water with 0.05% TFA) R_f0.63; H-NMR (500 MHz,

CF₃COOD) δ 1.47–1.55 (m, 2H), 1.73 (m, 2H), 1.79–1.95 (m,

2H), 2.02–2.69 (m, 4H), 2.43–2.62 (m, 4H), 2.68 (t, 2H, J =

6.7), 2.75 (t, 2H, J = 6.7), 3.03 (s, 4H), 3.23 (s, 6H), 3.38 (m,

2H), 3.45–3.56 (m, 2H), 3.61 (br t, 2H), 3.69–3.77 (m, 4H),

3.85 (m, 2H), 3.93 (m, 2H), 4.52 (br t, 1H), 6.88 (s, 2H); MAL-

DI-TOF MS m/z 717.1 (M + H)⁺; HRMS m/z 715.2954

(M + H)⁺(715.2973 calculated).

Compound ix. Compound vii (200 mg, 0.35 mmol) was added

to a mixture of MeOH and H₂O (95/5, 20 mL). Then, 10% of

palladium on carbon (20 mg) was added. The resulting suspen-

sion was hydrogenated at room temperature using a balloon for

4 h. The reaction mixture was then ﬁltered. The ﬁltrate was con-

centrated under reduced pressure at room temperature. The

desired product, compound viii, was obtained as a clear oil and

was used as such for the next reaction. Yield = 176 mg (quanti-

5. Hydrophobic peptide labeling (Fig. 2 and Figures S12–

S14, supplementary material†) with zwitterionic reagent 1. The

following is an illustrative procedure for the labeling of penta

(phenylalanine) with compound 1, as shown in Fig. 2. A sol-

ution of penta(phenylalanine) (2 mg, 2.65 μmol, Aldrich) in

DMSO (0.8 mL) was treated with compound 1 (2 mg, 4.8 μmol)

dissolved in DMSO (0.2 mL) followed by aqueous sodium car-

bonate (0.2 mL, 100 mM, pH 9). The reaction was stirred at

room temperature for 16 h. HPLC analysis of a small portion of

the reaction mixture was performed using a Phenomenex,

10 micron, C₁₈, 3.9 × 300 mm column and a 30 min gradient of

10 → 100% MeCN/water (with 0.05% TFA) at a ﬂow rate of

1.0 mL min⁻¹and UV detection at 220 and 260 nm. The labeled

peptide 6 was observed to elute at 18.5 min. The crude reaction

mixture was puriﬁed by HPLC using a YMC, 10 micron, C₁₈,

30 × 250 mm column and a 30 min gradient of 10 → 100%

MeCN/water (with 0.05% TFA) above at a ﬂow rate of 20 mL

min⁻¹. HPLC fractions containing product were frozen at

−80 °C and lyophilized to dryness to afford a ﬂuffy white solid.

Yield = 2.4 mg (92%); MALDI-TOF MS m/z 988.4 M⁺(988.4

calculated).

tative); MALDI-TOF MS m/z 440.3 (M

+

H)⁺(439.3

calculated).

A solution of compound viii (176 mg, 0.4 mmol) in anhydrous

DMSO (4 mL) was treated with 4-maleimidobutyric acid N-

hydroxsuccinimide ester (50 mg, 0.178 mmol, Thermo) and tri-

ethylamine (0.054 mL, 0.4 mmol). The reaction was stirred at

room temperature. After 1–2 h, HPLC analysis was performed

using a Phenomenex, 10 micron, C₁₈, 3.9 × 300 mm column and

a 40 min gradient of 0 → 40% MeCN/water (with 0.05% TFA)

at a ﬂow rate of 1.0 mL min⁻¹and UV detection at 260 &

220 nm. Product was observed eluting at 27.5 min. The crude

reaction mixture was puriﬁed by HPLC using a YMC,

10 micron, C₁₈, 30 × 250 mm column and a 40 min gradient of

0 → 40% MeCN/water (with 0.05% TFA) at a ﬂow rate of

20 mL min⁻¹. The HPLC fractions containing product were

frozen at −80 °C and lyophilized to dryness. Yield = 45 mg

(40%); HPTLC (3 : 7, MeCN : water with 0.05% TFA) R_f0.59;

¹H-NMR (500 MHz, CF₃COOD) δ 1.59 (s, 9H), 1.74 (br t, 2H),

2.07 (br t, 4H), 2.44 (m, 2H), 2.62 (br t, 2H), 3.23 (s, 3H), 3.27

(s, 3H), 3.37 (m, 2H), 3.51 (m, 2H), 3.61–3.82 (m, 8H), 4.28

(m, 1H), 4.43 (m, 1H), 6.88 (s, 2H); MALDI-TOF MS m/z

604.9 (M + H)⁺; HRMS m/z 604.3013 (M + H)⁺(604.3016

calculated).

Compound 5. A solution of compound ix (25 mg, 41 μmol) in

triﬂuoroacetic acid (2 mL) was stirred in an ice-bath for 3 h.

Anhydrous ether (75 mL) was then added to the reaction mixture

to precipitate the product which was collected by ﬁltration and

rinsed with ether. The product was then dissolved in methanol

(15–20 mL) and analyzed by HPLC as described above.

Product, compound x, was observed eluting at 13.7 min (com-

plete conversion). The methanol solution was diluted with anhy-

drous toluene (10 mL) and was concentrated under reduced

The other two peptides Phe-Gly-Gly-Phe and penta(leucine)

were labeled in a similar manner.

Compound 7, Yield = 3.6 mg (quantitative); MALDI-TOF

MS m/z 660.8 M⁺(661.3 calculated).

Compound 8, Yield = 1.2 mg (43%); MALDI-TOF MS m/z

820.4 (M + H)⁺(819.5 calculated).

Aqueous solubility of the labeled peptides was estimated by

dissolving the lyophilized, zwitterion-labeled peptides 6, 7 and 8

in small quantities of de-ionized water (0.2 mL portions) at room

temperature followed by vortexing. Concentrations that afforded

homogeneous solutions were used to estimate the aqueous solu-

bility of the zwitterion-labeled peptide.

6. Synthesis of cross-linked peptide 9 from conjugation of

penta(phenylalanine) to the tetrapeptide Phe-Gly-Gly-Phe

(Fig. 3 and Figure S15, supplementary material†) using zwitter-

ionic, homobifunctional cross-linker 4

1892 | Org. Biomol. Chem., 2012, 10, 1883–1895

View Article Online

Table 2 General procedure for the labeling of BSA and BGG with

acridinium ester and zwitterion label 1

Compound xi. A solution of compound 4 (4.5 mg, 8.3 μmol)

in anhydrous DMSO (2 mL) was added to penta(phenylalanine)

(2 mg, 2.7 μmol) along with diisopropylethylamine (2 μL,

11.4 μmol). The reaction was stirred at room temperature. After

16 h, HPLC analysis of the reaction mixture was performed

using a Phenomenex, 10 micron, C₁₈, 3.9 mm × 300 m column

and a 30 min gradient of 10 → 100% MeCN/water (with 0.05%

TFA) at a ﬂow rate of 1.0 mL min⁻¹and UV detection at 220

and 260 nm. Product xi was observed eluting at 17 min. The

crude reaction mixture was puriﬁed by HPLC using a YMC,

10 micron, C₁₈, 30 × 250 mm column and a 30 min gradient of

10 → 100% MeCN/water (with 0.05% TFA) at a ﬂow rate of

20 mL min⁻¹. HPLC fractions containing product were frozen at

−80 °C and lyophilized to dryness to afford a ﬂuffy white solid.

Yield = 2.0 mg (64%); MALDI-TOF MS m/z 1187.8 (M + H)⁺

(1187.5 calculated).

Zwitterion reagent 1,

input (eq.)

# of zwitterion

labels

Conjugate

BSA-NSP-DMAE 10

BSA-NSP-DMAE 20

BGG-NSP-DMAE 10

BGG-NSP-DMAE 20

5

8

9

13

acridinium abbreviated as NSP-DMAE-NHS⁷ester (see Table 1

for structure) added as a solution in DMF (0.072 mL of a 2 mg

mL⁻¹solution in DMF). The reaction was stirred at room temp-

erature for 4 h. The reaction was then transferred to a 4 mL

amicon ﬁlter (MW 30 000 cutoff) and diluted with 3 mL de-

ionized water. The ﬁlter was centrifuged at 4000g for 8 min to

reduce the volume to ∼0.25 mL. This process was repeated two

more times. The ﬁnal conjugate was diluted to 0.8 mL with de-

ionized water to give a 5 mg mL⁻¹solution. Acridinium ester

label incorporation, measured by MALDI-TOF mass spec-

troscopy using sinnapinic acid as matrix indicated the incorpor-

ation of 1.4 labels (Figure S17, supplementary material†).

Further labeling of the acridinium ester-labeled BSA conju-

gate with the zwitterion label 1 was carried out as follows. The

conjugate (0.5 mg, 7.5 nmol, 0.100 mL) was diluted with

0.100 mL of sodium carbonate, pH 9. To this conjugate was

added 10 and 20 equivalents of the zwitterion label 1 dissolved

in DMSO corresponding to 0.0128 mL and 0.0256 mL of a

2.5 mg mL⁻¹solution of 1. The reactions were stirred at room

temperature for 4 h and were then processed as described above.

Zwitterion label incorporation was measured by MALDI-TOF

mass spectroscopy and the results are listed in Table 2.

(Figure S17, supplementary material†)

Compound 9. A mixture of compound xi (2 mg, 1.68 μmol),

tetrapeptide Phe-Gly-Gly-Phe (3.6 mg, 5 equivalents) and diiso-

propylethylamine (1.5 μL, 5 equivalents) in DMSO (2.5 mL)

was stirred at room temperature. After 1 h, HPLC analysis was

performed as described above. Product 9 was observed eluting at

18 min. The crude reaction mixture was puriﬁed by HPLC using

a YMC, 10 micron, C₁₈, 30 × 250 mm column and a 30 min gra-

dient of 10 →100% MeCN/water (with 0.05% TFA) at a ﬂow

rate of 20 mL min⁻¹. HPLC fractions containing product were

frozen at −80 °C and lyophilized to dryness. Yield = 1.6 mg

(63%); MALDI-TOF MS m/z 1500.2 M⁺(1512.7 calculated).

7. Synthesis of cross-linked peptide 10 from conjugation of

penta(phenylalanine) with dipeptide Cys-Gly (Fig.

4 and

Figure S16, supplementary material†) using zwitterionic, hetero-

bifunctional cross-linker 5. A solution of penta(phenylalanine)

(8 mg, 11.2 μmol) and compound 5 (4 mg, 5.6 μmol) in anhy-

drous DMF (3 mL) was treated with triethylamine (4 μL, 4

equivalents). The reaction was stirred at room temperature for

16 h and then analyzed by HPLC using a Phenomenex,

10 micron, C₁₈, 3.9 × 300 mm column and a 30 min gradient of

10 → 100% MeCN/water (with 0.05% TFA) at a ﬂow rate of

1.0 mL min⁻¹and UV detection at 220 and 260 nm. Product xii

was observed eluting at 18.4 min (MALDI-TOF MS m/z 1355.8

[M + H]⁺; 1353.6 calculated). Starting material 5 eluting at

10 min was completely consumed. To the reaction mixture was

added the dipeptide Cys-Gly (2 mg, 11.1 μmol). The reaction

was stirred at room temperature. After 20 min, HPLC analysis

indicated clean formation of the heptapeptide 10 eluting at

16.5 min with no xii at 18.4 min. The heptapeptide 10 was

puriﬁed by HPLC using an YMC, 10 micron, C₁₈, 30 × 250 mm

column and a 30 min gradient of 10 → 100% MeCN/water (with

0.05% TFA) at a ﬂow rate of 20 mL min⁻¹. HPLC fractions con-

taining product were frozen at −80 °C and lyophilized to

dryness. Yield = 8.8 mg (quantitative) MALDI-TOF MS m/z

1535.3 (M + H)⁺(1531.7 calculated).

9. General procedure for the labeling of AVD, BGG and

FBN with acridinium ester and zwitterion label 1 for measure-

ment of non-speciﬁc binding. A solution of 2 mg mL⁻¹of egg

white a

̲

f

̲

bonate, pH 9. The protein solutions were then treated with 5

equivalents of the acridinium ester NSP-DMAE-NHS,⁷(Table 1)

added as a solution in DMSO. Speciﬁcally, a 2 mg mL⁻¹sol-

ution of the acridinium ester in DMSO was prepared and

0.089 mL, 0.04 mL and 0.018 mL of this solution were added to

AVD, BGG and FBN respectively. The reactions were stirred at

4 °C for 3 h. A portion (0.5 mL) of each reaction was further

treated with a 25 equivalents of compound 1 added as a solution

in DMSO. Speciﬁcally, a 5 mg mL⁻¹solution of 1 in DMSO

was prepared and 0.032 mL, 0.014 mL and 0.006 mL of this sol-

ution were added to acridinium ester-labeled AVD, BGG and

FBN respectively. The reactions were stirred at 4 °C for 16 h. All

three conjugates were then diluted with 0.5 mL each of 0.1 M

phosphate, pH 7.4 and transferred to 4 mL centricon ﬁlters. The

conjugates were further diluted with 3 mL buffer and centrifuged

at 4000g for 7 min to reduce the volume to ∼0.2 mL. This

process was repeated two more times. The ﬁnal conjugates were

diluted to a total volume of 0.5–1.0 mL in phosphate buffer and

stored at 4 °C.

8. General procedure for the labeling of BSA and BGG with

acridinium ester and zwitterion label 1. The following pro-

cedure was typical. A solution of BSA (4 mg, 60 nmol) in 1 mL

of 0.1 M sodium bicarbonate was treated with 5 equivalents of

the acridinium ester, 9-[[4-[[(2,5-dioxo-1-pyrrolidinyl)oxy]car-

bonyl]-2,6-dimethylphenoxy]carbonyl]-10-(3-sulfopropyl)-

Org. Biomol. Chem., 2012, 10, 1883–1895 | 1893

View Article Online

10. Measurement of FNSB (Table 1). The FNSB of three

proteins AVD, BGG and FBN, each labeled with the acridinium

ester NSP-DMAE and the zwitterion label 1, were measured and

compared to the FNSB of the three same proteins, labeled only

with the acridinium ester. The conjugates were diluted to equiv-

alent concentrations of 10 nM in a solution consisting of 0.1 M

sodium N-(2-hydroxyethyl) piperazine-N′-2-ethanesulfonate

(HEPES), 0.15 M sodium chloride, 7.7 mM sodium azide,

1.0 mM tetrasodium ethylenediaminetetraacetatic acid (EDTA),

12 mM t-octylphenoxypolyethoxyethanol (Triton X-100),

0.076 mM BSA and 0.007 mM mouse immunoglobin, pH 7.7.

0.1 mL of these 10 nM protein–acridinium ester conjugate sol-

utions were each mixed with 0.2 mL of horse serum and 0.2 mL

of either of three solid phases. The ﬁrst solid phase was 0.35 g

L⁻¹paramagnetic microparticles (PMP) covalently covered with

sheep antibody. The second solid phase was 0.35 g L⁻¹of Invi-

trogen Corporation M-280 magnetic latex microparticle (MLP)

Dynalbeads covalently covered with streptavidin. The third solid

phase was 0.35 g L⁻¹of Invitrogen Corporation M-270 magnetic

latex microparticle (MLP) Dynalbeads covalently covered with

streptavidin. The solid phases were magnetically collected and

washed twice with water after an incubation of 10 min to allow

interaction between the acridinium ester covalently attached pro-

teins and the solid phases. The chemiluminescence (expressed as

magnetically collected and washed with water to remove

unbound acridinium ester conjugate. Chemiluminescence values

(expressed as RLUs) corresponding to each theophylline stan-

dard concentration were normalized to the percentage of chemi-

luminescence measured for the zero standard (Fig. 5, panel A).

The amount of theophylline in a standard is inversely correlated

to the amount of the acridinium ester–theophylline conjugate

that will bind to the solid phase in the assay and likewise

inversely correlated to the resulting chemiluminescence. The

ratio of the bound chemiluminescence of the acridinium ester–

theophylline conjugate to the solid phase (B₀) in the absence

of unlabelled theophylline to the total chemiluminescence (T)

was expressed as a percentage (B₀/T%). Assay sensitivity

was assessed as the lowest theophylline concentration whose

chemiluminescence was lower than two-standard deviations of

the chemiluminescence for the zero standard. FNSB was

measured as the ratio of the chemiluminescence corresponding

to a very high concentration of theophylline to the total

chemiluminescence.

12. TSH (Fig. 6 and Table S2, supplementary material†).

Likewise, anti-TSH monoclonal antibody conjugates of the

labels compound 13⁷and compound 14¹⁷(Fig. 6, panel B) were

compared in the TSH assay. Conjugates (4 labels per antibody

for both labels) were prepared as described previously^7aand

were diluted to a concentration of 2.2 nM in a buffer of 0.10 M

HEPES, 0.15 M sodium chloride, 7.7 mM sodium azide,

1.0 mM EDTA, 12 mM triton X-100, 76 μM BSA, 97 nM

amphotericin B, 5.2 μM gentamicin sulfate, 6.7 μM murine IgG

(mouse immunoglobulin G), 67 nM sheep IgG, 5.3 mg L⁻¹

murine serum, 50 mg L⁻¹Antifoam B (Sigma Cat. No. A5757)

pH 7.7. The automated immunochemistry analyzer performed

the following steps for the TSH assay. First, 0.20 ml of each of

12 serum standards of known amounts of TSH was dispensed

into separate cuvettes. These standards contained respectively 0,

0.002, 0.004, 0.010, 0.015, 0.020, 0.025, 0.030, 0.10, 1.0, 10

and 100 mIU L⁻¹concentrations of TSH (1 IU = 1.58 × 10⁻⁸

mol). The instrument then dispensed 0.10 mL of 2.2 nM anti-

TSH antibody–acridinium ester conjugate solution to each

cuvette. The assay then proceeded for 5.0 min at 37 °C. To this

was added 0.225 mLTSH assay solid phase, containing magneti-

cally separable PMPs covalently derivatized with sheep anti-

TSH polyclonal antibody. The assay proceeded for another

5.0 min at 37 °C after which the solid phase was magnetically

collected and washed with water to remove unbound acridinium

ester conjugate. Chemiluminescence (expressed as RLUs) was

determined for each TSH standard. The assay results are illus-

trated in Fig. 6, Panel A and Table S2 (supplementary material†).

The amount of TSH in a standard is correlated to the amount of

anti-TSH antibody–acridinium ester conjugate that will bind to

the solid phase in the assay and consequently to the resulting

chemiluminescence.

RLUs, R

̲

chemiluminescence and the chemiluminescence of the acridi-

nium ester associated with the particles was measured for 5 s on

a Berthold Technologies’ Autolumat LB953 luminometer with

sequential addition of 0.3 mL each of reagent 1 (0.1 M nitric

acid and 0.5% hydrogen peroxide) and reagent 2 (0.25 M

sodium hydroxide containing 7 mM cetyltrimethylammonium

chloride). FNSB was calculated as the ratio of particle-bound

chemiluminescence from acridinium ester bound to the solid

phases divided by the total chemiluminescence of acridinium

ester input.

Invitrogen product literature descriptions of M-270 and

M-280 particles: M-270: Hydrophilic bead surface based on gly-

cidyl ether (epoxy) carboxylic acid beads; diameter: 2.8 μm;

M-280: Hydrophobic bead surface based on polystyrene tosyl

activated beads; diameter: 2.8 μm.

11. Theophylline assay (Fig. 5 and Table S1, supplementary

material†). Immunoassay properties including assay sensitivity,

FNSB and precision were assessed for acridinium ester–theo-

phylline conjugates 11 and 12^7,17(Fig. 5, panel B, Table S1, sup-

plementary material†) using a Siemens Healthcare Diagnostics’

ADVIA: Centaur^®system, an automated immunoanalyzer. Con-

jugates 11 and 12 were diluted to concentrations of 0.26 nM in

0.05 M sodium phosphate buffer pH 7.4 containing 0.15 M

sodium chloride, 1.1 mM EDTA, 0.22 mM triton X-100,

0.023 mM BSA, and 15 mM sodium azide. The immunoanaly-

zer automatically dispensed 0.020 mL each of ten serum stan-

dards containing 0, 1, 2, 4, 10, 20, 40, 100, 500, and 1000 μM

theophylline into reaction cuvettes. Then 0.45 mL of theophyl-

line assay solid phase, containing magnetically separable, PMPs

covalently linked with anti-theophylline antibody, was dispensed

to each cuvette. To this was added 0.10 mL of 0.26 nM theo-

phylline–acridinium ester conjugate solution. The assay pro-

ceeded for 7.5 min at 37 °C after which the solid phase was

For both assays standard curves were ﬁtted using four par-

ameter logistic (4PL) splines. Precision was measured as stan-

dard deviation of chemiluminescence or the standard deviation

relative to the mean chemiluminescence as a percentage. Func-

tional sensitivity for both assays is deﬁned here as the lowest

measured theophylline or TSH concentration distinguishable by

at least two standard deviations from the zero concentration-

1894 | Org. Biomol. Chem., 2012, 10, 1883–1895

View Article Online

G. Meluleni, E. Kim, L. Yan, G. Pier, H. S. Warren and G.

M. Whitesides, Langmuir, 2001, 17, 1225–1233; (c) S. J. Soﬁa,

V. Premnath and E. W. Merrill, Macromolecules, 1998, 31, 5059–5070;

(d) Y.-H. Zhao, B.-K. Zhu, L. Kong and Y.-Y. Xu, Langmuir, 2007, 23,

5779–5786; (e) A. R. Denes, E. B. Somers, A. C. L. Wong and F. Denes,

J. Appl. Polym. Sci., 2001, 81, 3425–3438.

standard. Enhancement of functional sensitivity is a decrease in

the lowest detectable TSH concentration as a result of using one

kind of acridinium ester label relative to another. FNSB is

deﬁned here as the ratio of the chemiluminescent signals associ-

ated with non-speciﬁc binding versus speciﬁc signal in the assay

(ratio of the chemiluminescence corresponding to the 0 mIU L⁻¹

TSH standard divided by the total chemiluminescence input per

assay).

10 R. G. Laughlin, Langmuir, 1991, 7, 842–847.

11 R. E. Holmlin, X. Chen, R. G. Chapman, S. Takayama and G.

M. Whitesides, Langmuir, 2001, 17, 2841–2850.

12 (a) J. Ladd, Z. Zhang, S. Chen, J. C. Hower and S. Jiang, Biomacromole-

cules, 2008, 9, 1357–1361; (b) Y. Chang, S. Chen, Z. Zhang and

S. Jiang, Langmuir, 2006, 22, 2222–2226; (c) Z. Zhang, T. Chao,

S. Chen and S. Jiang, Langmuir, 2006, 22, 10072–10077; (d) W. K. Cho,

B. Kong and I. S. Choi, Langmuir, 2007, 23, 5678–5682; (e) G. Cheng,

Z. Zhang, S. Chen, J. D. Bryers and S. Jiang, Biomaterials, 2007, 28,

4192–4199; (f) H. Kitano, A. Kawasaki, H. Kawasaki and S. Morokoshi,

J. Colloid Interface Sci., 2005, 282, 340–348.

References

1 (a) L. J. Kricka, Anal. Chim. Acta, 2003, 500, 279–286; (b) A. Roda and

M. Guardigli, Anal. Bioanal. Chem., 2012, 402, 69–76.

2 E. Y. Chi, S. Krishnan, T. W. Randolph and J. F. Carpenter, Pharm. Res.,

2003, 20, 1325–1336.

3 S. S. Levinson and J. J. Miller, Clin. Chim. Acta, 2002, 325, 1–15.

4 (a) N. C. Pomroy and C. M Deber, Bioconjugate Chem., 1999, 275, 224–

230; (b) A. Basu, K. Yang, M. Wang, S. Liu, R. Chintala, T. Palm,

H. Zhao, P. Peng, D. Wu, Z. Zhang, J. Hua, M. C. Hsieh, J. Zhou,

G. Petti, X. Li, A. Janjua, M. Mendez, J. Liu, C. Longley, Z. Zhang,

M. Mehlig, V. Borowski, M. Viswanathan and D. Filpula, Bioconjugate

Chem., 2006, 17, 618–630; (c) Y. Marsac, J. Cramer, D. Olschewski,

K. Alexandrov and C. F. W. Becker, Bioconjugate Chem., 2006, 17,

1492–1498.

13 Z. G. Estephan, P. S. Schlenoff and J. B. Schlenoff, Langmuir, 2011, 27,

6794–6800.

14 (a) M. E. Goldberg, N. Expert-Bezançon, L. Vuillard and T. Rabilloud,

Folding Des., 1995, 1, 21–27; (b) L. Vuillard, D. Madern, B. Franzetti

and T. Rabilloud, Anal. Biochem., 1995, 230, 290–294; (c) N. Expert-

Bezançon, T. Rabilloud, L. Vuillard and M. E. Goldberg, Biophys.

Chem., 2003, 100, 469–479.

15 (a) T. Collins, S. D’Amico, D. Georlette, J. C. Marx, A. L. Huston and

G. Feller, Anal. Biochem., 2006, 352, 299–301; (b) S. D’Amico and

G. Feller, Anal. Biochem., 2009, 385, 389–391.

5 S. S. Pai, T. M. Przybycien and R. D. Tilton, Langmuir, 2010, 26,

18231–18238.

16 J. Xiao, A. Burn and T. J. Tolbert, Bioconjugate Chem., 2008, 19, 1113–

1118.

6 J. M. Harris, N. E. Martin and M. Modi, Clin. Pharmacokinet., 2001, 40,

539–551.

17 A. Natrajan, D. Sharpe and D. Wen, Org. Biomol. Chem., 2011, 9, 5092–

5103.

7 (a) A. Natrajan, D. Sharpe, J. Costello and Q. Jiang, Anal. Biochem.,

2010, 406, 204–213; (b) A. Natrajan, D. Sharpe and Q. Jiang, US Patent

No. 6,664,043 B2, 2003.

18 C. Hamon, K. Kuhn, A. Thompson, D. Reuschling and J. Schaefer, US

2007/0023628 A1, 2007.

19 Reported solubility of peptides based on a Sciﬁnder (Chemical Abstracts

Service) search.

8 A. A. A. Ismail, Clin. Chem., 2005, 51, 25–26.

9 (a) R. C. Chapman, E. Ostuni, L. Yan and G. M. Whitesides, Langmuir,

2000, 16, 6927–6936; (b) R. C. Chapman, E. Ostuni, M. N. Liang,

20 A. T. Marttila, K. J. Airenne, O. H. Laitinen, T. Kulik, E. A. Bayer,

M. Wilchek and M. S. Kulomaa, FEBS Lett., 1998, 441, 313–317.

Org. Biomol. Chem., 2012, 10, 1883–1895 | 1895

Article Doi

Zwitterionic reagents for labeling, cross-linking and improving the performance of chemiluminescent immunoassays

DOI: 10.1039/c2ob06807a

Source and publish data:

Authors:

Article abstract of DOI:10.1039/c2ob06807a

Full text of DOI:10.1039/c2ob06807a

Products guided by the article

R&D Labs maybe for 1349808-68-9

Relevant to this article

Hot Product