Penicillin-Enhanced Chemiluminescence of The

7/18/2019 Penicillin-Enhanced Chemiluminescence of The

http://slidepdf.com/reader/full/penicillin-enhanced-chemiluminescence-of-the 1/3

Penicil l in-Enh anced Chemilum inescence of the

Luminol-H202-Co2+System

S. CHEN, .

YAN ,

M.

A.

SCHWARTZ,

.

H.

P E R R IN ,ND S.

G.

SCHULMAN'

Received November 6, 1990, from the College of Pharmacy, University o Florida, Gainesville, L 32670.

January 16, 1991.

Accepted for publication

Abstract

The I u ~ ~ ~ o I - H , O , - C O ~ +ystem has been widely used in

chemical and biological analysis. We report here an investigation of the

observation that penicillins have the ability to prolong and enhance the

intensity

of

chemiluminescence from luminol. The basis of this phenom-

enon appears, as revealed

by

differencespectroscopy, to be the

formation of a complex between the p-lactam and the superoxide ion.

The latter is the oxidizing species responsible for the oxidationof luminol

in

alkaline solution and has a mean lifetime,

n

solution, of milliseconds.

The stabilization of the superoxide

ion by

penicillin complexation extends

the effective lifetime of the superoxide ion by a few orders of magnitude

and thereby allows for more efficient oxidation of the p-lactam. Several

penicillins were determined by their enhancement of luminol chemilu-

minescence. A detection

limit

of 100

ng

mL was obtained for penicillin G

with a less-than-ideal detection system.

The chemiluminescence system luminol-H202-Co2+ LHC)

has been widely used in chemical and biological systems and

has been applied to pharmaceutical analysis.' The origin of

luminescence in this system can be briefly described as the

interaction of luminol, under alkaline conditions, with oxy-

gen to form an adduct which decomposes into N, and excited

aminophthalate. The la tte r species fluoresces. The Co2+ s a

catalyst which accelerates the decomposition of H,O, t o

supply the reactive form of oxygen which is needed to interact

with luminol. There are two very reactive types of oxygen

which ar e able to exist in solution, one5 s singlet O,, the other

is *Oi,he superoxide iom6 Currently, there is considerable

experimental evidence to show that singlet oxygen is not

predominant in strong alkali ~olution.~he current experi-

ments support this conclusion. The only active oxygen species

which plays an important role under alkaline conditions is

0;;ts lifetime in alkali is only several milliseconds.' The

peroxide adduct formed between i t and luminol can exist for

-4

s. This is the reason why the sensitivity of chemilumi-

nescence is low, and the duration short. I t is proposed that if

one compound (in this case penicillin) can interact with -0 -

to form a peroxide adduct and the adduct can then interact

with luminol ( thereby freeing the compound), then the sen-

sitivity and duration of luminescence will be promoted. The

structure of penicillin has some similarities to that of luminol,

in particular a highly strained heterocyclic ring. Conse-

quently, it seems possible that penicillin can form peroxide

adducts with 0;. The current investigations show tha t the

sensitivity and duration of luminescence are enhanced by

adding penicillin to the LHC system. This phenomenon

should be useful for the determination of penicillins at

picomolar concentrations or lower, as might be desirable for

checking for contamination by penicillins in clean rooms.

Experimental Section

R e a g e n t e T h e H,Oz (30%) was supplied by Fisher Scientific

Company. Penicillin G an d penicillin V were purchased from Sigma

Chemical Company, St. Louis, MO. iperacillin-Na was from Amer-

ican Cyanam id, Pea rl River, NY. 3-Aminophthalhydrazide (luminol),

CoC1, hydr ate (99.99 ), an d 1,4-diazabicyclo 2,2,2)octaneDABCO)

were purchased from Aldrich Chemical Company Inc., Milwaukee,

WI. All reagents were used as supplied.

Instruments-Absorption an d fluorescence spectra were measured

on a double-beam U V-vis spectrophotometer (model Lamda-3B) and

fluorescence spectro meter (model LS-5,Perkin -Elm er, Norwalk, CT),

respectively. All spectroscopic measur emen ts w ere blank-corrected.

Absorbance meas urem ents were carried out in matched reference and

sample cells. Chemiluminescence was m easure d by a spectrofluorim-

eter (model MK-1; Farrand Optical Company, Valhalla, NY and

recorded on a Fisher Recordall series 5000 strip chart recorder (Fisher

Scientific, Pittsbu rg, PA).

Procedure for Measuring Detect ion Limits of Penicillin and

Co2+-A

0.001 M luminol solution was prepared in

0.1

M NaOH. A

0.001 M aqu eous solution of H,O, was used thro ughout the detection

limit measu reme nts. The CoCl, an d penicillin aqueous solutions

were prepared at concentrations ap prop riate for a given experiment.

The ra tio of solution volumes for luminol: H,Oz: Co2+:penicillin was

0.8:0.7:0.03:0.03hroughout.

The luminol, CoCl,, an d penicillin solutions were placed in the

cuve tte tha t was positioned i n the cell holder of the fluorimeter. The

excitation slit was closed and th e emission slit was opened

as

wide

as

possible. The emission monochromator was set

at

420 nm, the

wavelength max imum of luminol fluorescence. The cover of the cell

comp artme nt was replaced in pa rt by the phosphorescence accessory

which allowed th e additio n of th e H,Oz solution without exposure to

light. The dependence of chemiluminescence on time was then

recorded. The peak heigh t a nd a rea u nder th e decay curve were used

to measure t he sensitivity. Both gave consistent results. When the

detection limit of Co2+ was dete rmin ed, the co ncentration of penicil-

l in was kept constant a t

-1

x M. When th e detection limit of

penicillin w as measured, the Co2+concentration was kept constant at

-1.9

x

M.

ll of the measurements were made at room

temperature. For each sample,

at

least three measurements were

made with good reproducibility.

Results and Discussion

Chemiluminescence Promoted b y Penicillin-The

chemiluminescence of luminol-H,0,-Co2+ and luminol-

H,02-Co2+-penicillin were measured under the same condi-

tions.

As

shown in Figure 1, the behaviors of these two

systems are quite different. For th e luminol-H,0,-Co2+ sys-

tem, only a brief pulse of light was observed, but on the

addition of penicillin, the sensitivity is much improved and

time of luminescence lengthened. The dependence of chemilu-

minescence intensity on the concentrations of luminol, Co2+,

H 02

nd penicillin has been studied. The linear relationship

between them allows chemiluminescence to be used for

quantitative measurements of Co2+ and penicillin.

Measurement of Detection Limits

of

Co2+ and Penicil-

lin-The concentration of luminol, H,O,, and penicillin were

kept constant and the Co2+ concentration varied. Chemilu-

minescence intensity-time curves were obtained, and at a

range of

lo-'

to g/mL, a str aight line was obtained with

a 0.99 linear correlation coefficient. Reproducibility was very

good. At this concentration, the noise was very small

so

the

OO22-3549/9 1100 7 02.50/0

0

799

1

American Pharmaceutical Association

Journal of Pharmaceutical Sciences 1017

Vol. SO,No.

11,

November 1991



Time min)

Flgure1-Dependence of chemiluminescenceon time. Key:

(A)

luminol-

H202-Co2+;B) luminol-H202-Co2+-penicillin.

The concentrations were

as follows: penicillin,

9.4 x

M;

H20 2, 4.4 x

M; and luminol, 5.0 x

M;

Co2 , 1.9 x

M in

0.05 M

NaOH solution.

detection limit was 6

x lo-

g/mL or lower. This is in spite

of

the fact tha t the optical system was not ideal because the

cuvette is a t least

60

cm from the photomultiplier as multiple

reflections are used in the emission optics of the Farrand

spectrofluorimeter. Usual chemiluminescence instruments

have a sample-to-detector distance of only -2 cm. The

concentration ratios of the reactants probably was not opti-

mal, giving further opportunity for increasing the sensitivity.

Using constant concentrations of luminol, H202, and Co2+,

the detection limit of penicillin

G

was found to be 100 ng/mL.

Effect

of

Different Penicillins on the Catalytic Activity-

Penicillin G, Penicillin V, and pipericillin promoted the

chemiluminescence to varying extents. Penicillin G and

pipericillin had almost the same effect, while that of penicillin

V was a little smaller.

Reaction Mechanism-The cobalt ion (Co2+) s a catalyst

for the luminol-H,O, system. However, penicillin alone is not

a catalyst because in an alkaline solution of luminol-H,O,-

penicillin without Co2+,chemiluminescence is not observed.

With the concentrations of luminol, H,02, and penicillin

fixed and the concentration of Co2+ varied, two types of

luminescence curves were observed. When the ratio

Co2+:penicillinwas high, the intensity of chemiluminescence

increased rapidly to a maximum, then i t decayed following a

single-decay exponential. On the other hand, when the ratio

was low, the chemiluminescence intensity increased rapidly

t o

a maximum, then it decreased rapidly subsequently,

increasing slowly to a second maximum before decreasing

again to zero, as shown in Figure

2.

These experimental

results indicate that the reaction should be divided into two

parts. The rapid increase and decrease is contributed by the

luminol-H,0,-Co2+ system and the slower part is derived

from the role of penicillin added to the solution.

As is well known, the chemiluminescence reaction mecha-

nisms can be briefly expressed by the following reaction:

Luminol

+

[Oxygen+Peroxide Adduct-

Aminophthalate-*Aminophthalate

hv

1)

In solution, there are two forms of oxygen which may play

an important role in reacting with luminol. One is singlet

oxygen and the other is th e superoxide anion C O i ) . The very

effective singlet oxygen scavenger DABC06*' was added t o

the luminol-H20~+-penicillinystem, and no observable

change in the chemiluminescence was found. Previous work

has shown that singlet oxygen is not stable in alkaline

s ~ l u t i o n . ~herefore, it is concluded that the role of singlet

oxygen can be excluded and the superoxide anion is probably

the only species involved in the oxidation of luminol to

aminophthalate.

Time (min)

Figure 2-Dependence of chemiluminescenceon time.

The

concentra-

tions were the same as in Figure 1.

The lifetime of the superoxide anion is only several milli-

seconds: even in alkaline solution. Addition of penicillin to

the luminol-H,O,-Co2+ system extends the duration of

chemiluminescence to several minutes. This indicates that

penicillin can combine with the superoxide anion t o form a

peroxide adduct which prevents the decomposition of the

superoxide. The luminescent species is aminophthalate with

or without penicillin. This means that the penicillin peroxide

adduct can react with luminol to exchange superoxide anions.

The reaction mechanism can be expressed as follows:

luminol Oi-*luminol-peroxide adduct

luminol peroxide adduct-*aminophthalate

(2)

-+ aminophthalate + hv

penicillin Oi-tpenicillin-peroxide adduct (3)

penicillin-peroxide adduct

+

luminol+luminol-

peroxide adduct penicillin (4)

luminol-peroxide adduct-aminophthalate hv

( 5 )

A critical aspect of the above proposal is to find evidence of

the existence of a penicillin-peroxide adduct. As discussed

above, the lifetime of the superoxide anion is very short, but

the penicillin-peroxide adduct should have

a

relatively long

lifetime. An ordinary UV-vis spectrophotometer was used to

measure the absorption spectrum of the H,02.-Co2 +-penicillin

alkaline solution from 240 to

300

nm by using a penicillin-

Co2+ alkaline solution in the reference cell. At these wave-

lengths, in the absence of interaction in the sample cell, the

absorptions of penicillin in the reference and sample cells

should cancel. The superoxide anion has a very short lifetime,

and cannot be detected. If light is absorbed

at

these wave-

lengths,

it

should result from the penicillin-superoxide ad-

duct. The experimental result is shown in Figure

3.

This

101

8 Journal

of Pharmaceutical Sciences

Vol. 80,

No.

1 1

November 7991



B

0 1

2w 220 24 260

280 3w

Wavelength nm)

Figure +(A) Absorption spectrum of penicillin-peroxide adduct. The

concentrations were as follows: penicillin

G, 9.1

6 x 0 4 M;

o2 ,

1.83

x

M; NaOH,

4 6 x

M; and H,02,

4 76

x M. (B)

Absorption spectrum of 6 7

x

M penicillin G.

result confirms the existence of a penicillin-peroxide adduct.

The enhanced luminescence apparent from the penicillin-

LHC system is certainly of analytical value. However,

quenchers of 3-aminophthalate fluorescence (at concentra-

tions high enough to cause diffusional quenching: i.e. > 1 x

mol L- ) which include ions derived from elements

of

high atomic number and many aromatic molecules as well as

Lewis bases which ar e coordinated by Co2+ may interfere,

decreasing the analytical sensitivity. Consequently, the sys-

tem probably ought to be adapted to becoming the detector

ancillary to an efficient separation process such as HPLC or

f low injection analysis.

References and No tes

1.

Tsai, T. S., l in. Chem. 1985,31, 248.

2. Leupold,

C.;

olkl,A.;Fahimi, H. D.AnaZ.Biochem.1985,151,63.

3.

Nieman, T. A. Abstracts

of 3rd

International Symposium on

uantitative Lumineacence Spectrometry in Biomedical Sciences,

4.

Milbrawth, D.

S:

P m e e d i n g s o N t h International Bwlumines

5. Selinger, H H. hotochem. Photobiol. 1975, 73, 35.

6.

Rabani,

J.;

Niesen,

S.

0 J . Phys . Chem.

1969, 73, 736.

7. Ware, W. R.;Richter,

M.

P.

J Chem. Phys.

1968,48, 595.

8 .

Natl .

Stan dard Re f. Data Series; Ross F; Ross A.

B.

Eds.; Natl.

ent, Belgium, 1989, 34.

cence

and

ChemrluminescenceBmposaum. 1986; p 515-518.

Bur. Stand., 1977; 59.

Journal of Pharmaceutical Sciences

I

1019

Vol.

SO,

No. 7 7 November 7997

Penicillin-Enhanced Chemiluminescence of The

Documents

Transcript of Penicillin-Enhanced Chemiluminescence of The