- Email: [email protected]
An optical hydroxyl radical sensor
Biosensors & Bioelectronics 8 (1993) 325-329
An optical hydroxyl radical sensor Declan P. Naughton *, Martin Grootveld The Inflammation
& David R. Blake
Group, ARC Building, The London Hospital Medical College, London El 2AD, UK
Hector R. Guestrin & Ramaier Narayanaswamy Department
of Instrumentation
and Analytical
Science, UMIST, PO Box 88, Manchester
M60 lQD, UK
(Received 7 August 1992; revised version received 17 January 1993; accepted 27 January
1993)
Abstract: A hydroxyl
radical (*OH) fibre-optic sensor has been developed. An *OH radical-sensitive reagent phase (nitrophenol) was immobilized onto KAD-7 methacrylate beads. Subsequently the beads were attached to the distal end of a polymethylmethacrylate fibre optic. Nitrocatechol, generated from the attack of *OH radical on nitrophenol, exhibits a strong absorption band in the visible region of the electromagnetic spectrum ( Amru= 510 nm). Here, reflectance spectroscopy was employed to monitor the concomitant intensity decrease in the reflectance spectrum upon *OH radical attack. The sensor exhibited excellent stability and linearity of response to *OH generated by a Fenton reaction system (EDTA, Fe(I1) and HzO,) with H202 over the concentration range of 3.6 X lO+%O x 10m2 M. Keywords: Optical fibre, immobilized radical determination.
1. INTRODUCTION
Much interest has recently been shown in the detection of free radicals in view of the evidence implicating a primary or secondary role for them in the initiation or progression of many diseases, the majority of which are characterized by an inflammatory reaction (Halliwell & Gutteridge, 1989; Gutteridge & Stocks, 1981; Pippard et al., 1979). For the hydroxyl (*OH) radical, detection assays are based on spin-trap detection by electron spin resonance (ESR) spectroscopy, the degradation of dimethyl sulphoxide (DMSO), methionine and formate, and the hydroxylation and
* To whom correspondence
reagent,
nitrophenol,
sensor, hydroxyl
subsequent analysis of aromatic compounds (Buettner et al., 1978; Grootveld & Halliwell, 1986; Graf et al., 1984; Beauchamp & Fridovich, 1970; Sutton, 1985). The spectroscopic measurement of nitrocatechols, generated by *OH radical attack on nitrophenols, has been employed as a measure of *OH radical activity both in vitro and in vivo (eq. 1) (Buhler & Mason, 1961; Chrastil &
should be addressed.
0956-5663/93/$OfXNl0 1993 Elsevier Science Publishers Ltd.
325
D. P. Naughton
Biosensors & Bioelectronics
et al.
Wilson, 1975). Here we describe the development of a fibre-optic sensor, based on immobilized nitrophenol, that is of potential use for the continuous monitoring of *OH radical production. This reflectance-based sensor incorporates an aOH radical-sensitive chromophore, affording a decrease in its reflectance spectrum upon attack by this extremely reactive oxygen-derived radical attributable to the formation of nitrocatechol (es. 1).
2. EXPERIMENTAL 2.1.
Reagents
A number of nitrophenol analogues (nitrophenol, nitrosalicylate and nitrotyrosine) were tested for their suitability as reagent phase to be immobilized onto the distal end of the fibre-optic. The immobilization and *OH radical detection ability (once immobilized) of each reagent was tested on amberlite XAD-2, XAD4 and XAD7 beads in order to maximize sensitivity. The highest degree of sensitivity was found for the bead type XAD-7 with immobilized nitrophenol. The immobilization conditions for this reagent and bead type required immersion of the beads in a saturated aqueous (pH 7.0) solution of nitrophenol for a period of 2 h. 2.2. Probe construction The probe was constructed as previously described (Kirkbright et al., 1984; Alder et al., 1987). Briefly, the polished fibre tip (Radiospares, core 1 mm) was coated with nitrophenol-treated beads. A porous PTFE membrane (Millipore FHUP, pore size 0.5 pm, membrane thickness 60 pm) was used to encapsulate the probe tip. The PTFE membrane was held in place using heat-shrinkable tubing to which minimal heat was applied. The probe was sensitized by immersion in Analar grade methanol for 10 s to achieve membrane porosity. Once activated, the probe tip was stored in distilled and deionized water. A Fenton system containing EDTA, Fe(I1) and hydrogen peroxide was employed for the generation of hydroxyl radical (eq. 2) (Walling & Johnson, 1975). Fe2+ + H202 + Fe3+ + -OH + *OH 326
(2)
3. RESULTS
AND DISCUSSION
The reflectance-based fibre-optic sensor developed incorporates an *OH radical-sensitive chromophore affording a decrease in its reflectance spectrum upon attack by this radical. This chromophoric modification arises from the formation of nitrocatechol (eq. 1). The irreversible reaction facilitates continuous cumulative monitoring of *OH radical production. The spectral response of the reagent phase was measured by *OH generation using a Fenton system with various concentrations of added H202. The peak response attributable to the attack of *OH radical on the immobilized reagent was at a wavelength of 565 nm. The spectral response of the reagent phase was measured for various concentrations of added hydrogen peroxide. No response was detected in the absence of H202, or with H202 in the absence of Fe(I1) ions. As the aOH radical is an extremely short-lived species, and the EDTA-Fe(I1) system is an extremely powerful catalyst for H202 decomposition, kinetic measurements are difficult to make under these conditions. Hence, a cumulative measure of total *OH radical generation over time was conducted. Figure 1 shows a typical spectral response to increasing concentrations of *OH radical, generated by a Fenton system in aqueous solution (pH 5.0) over the concentration range of 2-O X 10-4-8.0 X 10-l M H202 (at a constant Fe(I1) ion concentration of 5 mrvr). The linearity of response of the sensor is demonstrated in plots of the decrease in reflectance at 565 nm as a function of added H202 over this concentration range. For measurements taken as a function of time, and for low concentrations of added H202, readings were taken at a single wavelength of 565 nm in order to overcome difficulties in plotting spectra. Typical response curves expressed as a function of time are given in Fig. 2. For the concentrations of H202 employed in this study a full response was achieved after 8 min. The time for probe response is characteristic of a diffusion-controlled process; i.e. increasing the concentration of H202 gave rise to an increase in time for probe response. The membrane was specifically chosen for its large pore size, which facilitates free diffusion of redox-active lowmolecular-mass iron species, and for its physical stability, a major requirement to ensure retention of the reagent phase. The probe response as a
Optical hydroxyl radical sensor
Biosensors & Bioelectronics
INTENSITY
INTENSITY
H?Ot
concw.aeion(mhq)
Fig. I. Sensor response to *OH radical generated by a Fenton system containing EDTA, Hz02 in the concentration range 2.0 x 10-4-8.0 x 10-l M.
INTENSITY
0
200
400
600
600
Time (set) Fig. 2. Sensor response to *OH radical generated by a Fenton system containing: (top curve) H,O,, 3.06 x IO-’ M, EDTA, and F2+, 8.29 x IO-4 M; (bottom curve) H202, 1,08 X IO+ M, EDTA, and Fti+t. 8.23 x 1O-4 M.
function of time, on addition of 366 X lop6 M HZ02, is plotted in Fig. 3. A linear decrease in reflectance as a function of the concentration of added H202 was observed down to a minimum concentration of 3.66 x lop6 M H202, and the reproducibility of the measured response is shown in Fig. 4. Under the experimental conditions
F$+,
5.0 x 10m3 M, and
good reproducibility was employed here, achieved. The lower limit of radical production detected was generated from a Fenton system containing only 3.66 X 1O-6 M H202. The sensor response as a function of pH was also investigated. These studies demonstrated that *OH radical detection varied minimally with pH throughout the range 4G8.0. The probe exhibited very good stability over the conditions employed in these experiments. It was designed as a disposable tool for clinical research and diagnostic purposes, and as such exhibited a good shelf-life stability (dry storage > 6 months) with an in-use lifetime dependent on the rate of generation of *OH radical within the system investigated.
4. CONCLUSIONS This is the first report, to our knowledge, of a fibre-optic based sensor suitable for the detection of hydroxyl radical (or iron-oxo species of comparable reactivity) production. The stability of the sensor, coupled to detection based on an irreversible reaction, allowed a cumulative response to be measured over extended time 327
D. P. Naughton et al.
Biosensors & Bioelectronics
periods. This sensor has potential application in the assessment of *OH radical production in cases of knee-joint and intestinal inflammatory diseases. Moreover, this probe may be employed to assess -OH radical production in vitro in chemical reaction vessels and in cell culture systems. Further work is currently underway to increase the sensitivity of this probe and to evaluate its performance characteristics in appropriate biological systems.
4
lNTENSlw
5.6g ;,
600
Time (set)
ACKNOWLEDGEMENTS We are grateful to the Arthritis and Rheumatism Council of Great Britain (DPN/MG/DRB) and to CNPq (Brazil) (HRG) for financial support, and to the family of Katharine Neilson for a generous donation.
Intensity Change
REFERENCES o.o!
.
0
,
.
20
1
.
I
40
.
I
60
.
,
60
IH2021
1
.
100
120
WV
Fig. 3. (A) Sensor response as a function of time to *OH radical generated by a Fenton system containing 5.0 x lO-3 M, and Hz02 in the EDTA, F$+, concentration range 366 x 10e6. (B) Sensor response to .OH radical generated by a Fenton system containing EDTA, Fe2+, 5.0 x IO-’ M, and H202 in the concentration range 3.66 x 10-6-1.08 X lop4 M.
INTENSITY
.*. ._-.. .....f.: i;$+ .”
.
. .
-.
.
. .
*
’
. ..
.
. . *
.-
5.795.76
Pharmacol. Exp. Theor.,
.
. _.
:’ . , : . . **..*. . .
243-50.
I_ 0
200
400
600
600
Time (set) Fig. 4. Reproducibility of sensor response to *OH radical generated by a Fenton system containing H202, 3.46 X low5 M, EDTA, and FeZ+, 8.29 X 10m4 M. 328
193, 631-8.
Graf, E., Mahoney, J.R., Bryant, R.G. & Eaton, J.W. (1984). Iron-catalysed hydroxyl radical formation. J. Biol. Chem., 259, 3620-4. Grootveld, M. & Halliwell, B. (1986). An aromatic hydroxylation assay for hydroxyl radicals utilising high performance liquid chromatography (HPLC). Use to investigate the effect of EDTA on the Fenton reaction. Free Rad. Res. Commun., l(4),
.*..
f.
. .. .
92, 424-37.
Chrastil, J. & Wilson, J.T. (1975). 4-Nitrocatechol production from p-nitrophenol by rat liver. J.
:
.
Alder, J.F., Ashworth, D.C., Narayanaswamy, R., Moss, R.E. & Sutherland, 1.0. (1987). An optical potassium ion sensor. Analyst, 112, 1191-2. Beauchamp, C. & Fridovich, I. (1970). Iron-catalysed hydroxyl radical formation. A mechanism for the production of ethylene from methional. J. Biol. Chem., 245, 4641-6. G.R., Uberley, L.W. & Leuthauser, Buettner, S.W.H.C. (1978). Effect of iron on the distribution of superoxide and hydroxyl radicals as seen by spin-trapping and on the superoxide dismutase assay. Photochem. Photobiol., 21, 693-S. Buhler, D.R. & Mason, H.S. (1961). Hydroxylation catalysed by peroxidase. Arch. Biochem. Biophys.,
Gutteridge, J.M.C. & Stocks, J. (1981). Caeruloplasmin: physiological and pathological perspectives. CRC Crit. Rev. Clin. Lab. Sci., 14, 257-329.
B. & Gutteridge, J.M.C. (1989). Free radicals in biology and medicine. Second edition,
Halliwell,
Clarendon
Press, Oxford.
Biosetwors & Bioelectronics Kirkbright, G.F., Narayanaswamy, R. &LWelti, N.A. (1984). Fibre optic pH probe based on the use of an immobilised calorimetric indicator. Analyst, 109,1025-8. F’ippard, M.J., Warner, G.T. Callender, S.T. & Weatherall, D.J. (1979). Iron absorption and loading in P-thalassaemia. Lancet ii, 819-21.
Optical hydroxyl radical sensor Sutton, H.C. (1985). Efficiency of chelated iron compounds as catalysts for the Haber-Weiss reaction. J. Free Rad. Biol. Med., 1, 195-202. Walling, C. & Johnson, R.A. (1975). Fenton’s reagent. V: Hydroxylation and side-chain cleavage of aromatics. J. Am. Chem. Sot., 97, 363-7.
329
An optical hydroxyl radical sensor Declan P. Naughton *, Martin Grootveld The Inflammation
& David R. Blake
Group, ARC Building, The London Hospital Medical College, London El 2AD, UK
Hector R. Guestrin & Ramaier Narayanaswamy Department
of Instrumentation
and Analytical
Science, UMIST, PO Box 88, Manchester
M60 lQD, UK
(Received 7 August 1992; revised version received 17 January 1993; accepted 27 January
1993)
Abstract: A hydroxyl
radical (*OH) fibre-optic sensor has been developed. An *OH radical-sensitive reagent phase (nitrophenol) was immobilized onto KAD-7 methacrylate beads. Subsequently the beads were attached to the distal end of a polymethylmethacrylate fibre optic. Nitrocatechol, generated from the attack of *OH radical on nitrophenol, exhibits a strong absorption band in the visible region of the electromagnetic spectrum ( Amru= 510 nm). Here, reflectance spectroscopy was employed to monitor the concomitant intensity decrease in the reflectance spectrum upon *OH radical attack. The sensor exhibited excellent stability and linearity of response to *OH generated by a Fenton reaction system (EDTA, Fe(I1) and HzO,) with H202 over the concentration range of 3.6 X lO+%O x 10m2 M. Keywords: Optical fibre, immobilized radical determination.
1. INTRODUCTION
Much interest has recently been shown in the detection of free radicals in view of the evidence implicating a primary or secondary role for them in the initiation or progression of many diseases, the majority of which are characterized by an inflammatory reaction (Halliwell & Gutteridge, 1989; Gutteridge & Stocks, 1981; Pippard et al., 1979). For the hydroxyl (*OH) radical, detection assays are based on spin-trap detection by electron spin resonance (ESR) spectroscopy, the degradation of dimethyl sulphoxide (DMSO), methionine and formate, and the hydroxylation and
* To whom correspondence
reagent,
nitrophenol,
sensor, hydroxyl
subsequent analysis of aromatic compounds (Buettner et al., 1978; Grootveld & Halliwell, 1986; Graf et al., 1984; Beauchamp & Fridovich, 1970; Sutton, 1985). The spectroscopic measurement of nitrocatechols, generated by *OH radical attack on nitrophenols, has been employed as a measure of *OH radical activity both in vitro and in vivo (eq. 1) (Buhler & Mason, 1961; Chrastil &
should be addressed.
0956-5663/93/$OfXNl0 1993 Elsevier Science Publishers Ltd.
325
D. P. Naughton
Biosensors & Bioelectronics
et al.
Wilson, 1975). Here we describe the development of a fibre-optic sensor, based on immobilized nitrophenol, that is of potential use for the continuous monitoring of *OH radical production. This reflectance-based sensor incorporates an aOH radical-sensitive chromophore, affording a decrease in its reflectance spectrum upon attack by this extremely reactive oxygen-derived radical attributable to the formation of nitrocatechol (es. 1).
2. EXPERIMENTAL 2.1.
Reagents
A number of nitrophenol analogues (nitrophenol, nitrosalicylate and nitrotyrosine) were tested for their suitability as reagent phase to be immobilized onto the distal end of the fibre-optic. The immobilization and *OH radical detection ability (once immobilized) of each reagent was tested on amberlite XAD-2, XAD4 and XAD7 beads in order to maximize sensitivity. The highest degree of sensitivity was found for the bead type XAD-7 with immobilized nitrophenol. The immobilization conditions for this reagent and bead type required immersion of the beads in a saturated aqueous (pH 7.0) solution of nitrophenol for a period of 2 h. 2.2. Probe construction The probe was constructed as previously described (Kirkbright et al., 1984; Alder et al., 1987). Briefly, the polished fibre tip (Radiospares, core 1 mm) was coated with nitrophenol-treated beads. A porous PTFE membrane (Millipore FHUP, pore size 0.5 pm, membrane thickness 60 pm) was used to encapsulate the probe tip. The PTFE membrane was held in place using heat-shrinkable tubing to which minimal heat was applied. The probe was sensitized by immersion in Analar grade methanol for 10 s to achieve membrane porosity. Once activated, the probe tip was stored in distilled and deionized water. A Fenton system containing EDTA, Fe(I1) and hydrogen peroxide was employed for the generation of hydroxyl radical (eq. 2) (Walling & Johnson, 1975). Fe2+ + H202 + Fe3+ + -OH + *OH 326
(2)
3. RESULTS
AND DISCUSSION
The reflectance-based fibre-optic sensor developed incorporates an *OH radical-sensitive chromophore affording a decrease in its reflectance spectrum upon attack by this radical. This chromophoric modification arises from the formation of nitrocatechol (eq. 1). The irreversible reaction facilitates continuous cumulative monitoring of *OH radical production. The spectral response of the reagent phase was measured by *OH generation using a Fenton system with various concentrations of added H202. The peak response attributable to the attack of *OH radical on the immobilized reagent was at a wavelength of 565 nm. The spectral response of the reagent phase was measured for various concentrations of added hydrogen peroxide. No response was detected in the absence of H202, or with H202 in the absence of Fe(I1) ions. As the aOH radical is an extremely short-lived species, and the EDTA-Fe(I1) system is an extremely powerful catalyst for H202 decomposition, kinetic measurements are difficult to make under these conditions. Hence, a cumulative measure of total *OH radical generation over time was conducted. Figure 1 shows a typical spectral response to increasing concentrations of *OH radical, generated by a Fenton system in aqueous solution (pH 5.0) over the concentration range of 2-O X 10-4-8.0 X 10-l M H202 (at a constant Fe(I1) ion concentration of 5 mrvr). The linearity of response of the sensor is demonstrated in plots of the decrease in reflectance at 565 nm as a function of added H202 over this concentration range. For measurements taken as a function of time, and for low concentrations of added H202, readings were taken at a single wavelength of 565 nm in order to overcome difficulties in plotting spectra. Typical response curves expressed as a function of time are given in Fig. 2. For the concentrations of H202 employed in this study a full response was achieved after 8 min. The time for probe response is characteristic of a diffusion-controlled process; i.e. increasing the concentration of H202 gave rise to an increase in time for probe response. The membrane was specifically chosen for its large pore size, which facilitates free diffusion of redox-active lowmolecular-mass iron species, and for its physical stability, a major requirement to ensure retention of the reagent phase. The probe response as a
Optical hydroxyl radical sensor
Biosensors & Bioelectronics
INTENSITY
INTENSITY
H?Ot
concw.aeion(mhq)
Fig. I. Sensor response to *OH radical generated by a Fenton system containing EDTA, Hz02 in the concentration range 2.0 x 10-4-8.0 x 10-l M.
INTENSITY
0
200
400
600
600
Time (set) Fig. 2. Sensor response to *OH radical generated by a Fenton system containing: (top curve) H,O,, 3.06 x IO-’ M, EDTA, and F2+, 8.29 x IO-4 M; (bottom curve) H202, 1,08 X IO+ M, EDTA, and Fti+t. 8.23 x 1O-4 M.
function of time, on addition of 366 X lop6 M HZ02, is plotted in Fig. 3. A linear decrease in reflectance as a function of the concentration of added H202 was observed down to a minimum concentration of 3.66 x lop6 M H202, and the reproducibility of the measured response is shown in Fig. 4. Under the experimental conditions
F$+,
5.0 x 10m3 M, and
good reproducibility was employed here, achieved. The lower limit of radical production detected was generated from a Fenton system containing only 3.66 X 1O-6 M H202. The sensor response as a function of pH was also investigated. These studies demonstrated that *OH radical detection varied minimally with pH throughout the range 4G8.0. The probe exhibited very good stability over the conditions employed in these experiments. It was designed as a disposable tool for clinical research and diagnostic purposes, and as such exhibited a good shelf-life stability (dry storage > 6 months) with an in-use lifetime dependent on the rate of generation of *OH radical within the system investigated.
4. CONCLUSIONS This is the first report, to our knowledge, of a fibre-optic based sensor suitable for the detection of hydroxyl radical (or iron-oxo species of comparable reactivity) production. The stability of the sensor, coupled to detection based on an irreversible reaction, allowed a cumulative response to be measured over extended time 327
D. P. Naughton et al.
Biosensors & Bioelectronics
periods. This sensor has potential application in the assessment of *OH radical production in cases of knee-joint and intestinal inflammatory diseases. Moreover, this probe may be employed to assess -OH radical production in vitro in chemical reaction vessels and in cell culture systems. Further work is currently underway to increase the sensitivity of this probe and to evaluate its performance characteristics in appropriate biological systems.
4
lNTENSlw
5.6g ;,
600
Time (set)
ACKNOWLEDGEMENTS We are grateful to the Arthritis and Rheumatism Council of Great Britain (DPN/MG/DRB) and to CNPq (Brazil) (HRG) for financial support, and to the family of Katharine Neilson for a generous donation.
Intensity Change
REFERENCES o.o!
.
0
,
.
20
1
.
I
40
.
I
60
.
,
60
IH2021
1
.
100
120
WV
Fig. 3. (A) Sensor response as a function of time to *OH radical generated by a Fenton system containing 5.0 x lO-3 M, and Hz02 in the EDTA, F$+, concentration range 366 x 10e6. (B) Sensor response to .OH radical generated by a Fenton system containing EDTA, Fe2+, 5.0 x IO-’ M, and H202 in the concentration range 3.66 x 10-6-1.08 X lop4 M.
INTENSITY
.*. ._-.. .....f.: i;$+ .”
.
. .
-.
.
. .
*
’
. ..
.
. . *
.-
5.795.76
Pharmacol. Exp. Theor.,
.
. _.
:’ . , : . . **..*. . .
243-50.
I_ 0
200
400
600
600
Time (set) Fig. 4. Reproducibility of sensor response to *OH radical generated by a Fenton system containing H202, 3.46 X low5 M, EDTA, and FeZ+, 8.29 X 10m4 M. 328
193, 631-8.
Graf, E., Mahoney, J.R., Bryant, R.G. & Eaton, J.W. (1984). Iron-catalysed hydroxyl radical formation. J. Biol. Chem., 259, 3620-4. Grootveld, M. & Halliwell, B. (1986). An aromatic hydroxylation assay for hydroxyl radicals utilising high performance liquid chromatography (HPLC). Use to investigate the effect of EDTA on the Fenton reaction. Free Rad. Res. Commun., l(4),
.*..
f.
. .. .
92, 424-37.
Chrastil, J. & Wilson, J.T. (1975). 4-Nitrocatechol production from p-nitrophenol by rat liver. J.
:
.
Alder, J.F., Ashworth, D.C., Narayanaswamy, R., Moss, R.E. & Sutherland, 1.0. (1987). An optical potassium ion sensor. Analyst, 112, 1191-2. Beauchamp, C. & Fridovich, I. (1970). Iron-catalysed hydroxyl radical formation. A mechanism for the production of ethylene from methional. J. Biol. Chem., 245, 4641-6. G.R., Uberley, L.W. & Leuthauser, Buettner, S.W.H.C. (1978). Effect of iron on the distribution of superoxide and hydroxyl radicals as seen by spin-trapping and on the superoxide dismutase assay. Photochem. Photobiol., 21, 693-S. Buhler, D.R. & Mason, H.S. (1961). Hydroxylation catalysed by peroxidase. Arch. Biochem. Biophys.,
Gutteridge, J.M.C. & Stocks, J. (1981). Caeruloplasmin: physiological and pathological perspectives. CRC Crit. Rev. Clin. Lab. Sci., 14, 257-329.
B. & Gutteridge, J.M.C. (1989). Free radicals in biology and medicine. Second edition,
Halliwell,
Clarendon
Press, Oxford.
Biosetwors & Bioelectronics Kirkbright, G.F., Narayanaswamy, R. &LWelti, N.A. (1984). Fibre optic pH probe based on the use of an immobilised calorimetric indicator. Analyst, 109,1025-8. F’ippard, M.J., Warner, G.T. Callender, S.T. & Weatherall, D.J. (1979). Iron absorption and loading in P-thalassaemia. Lancet ii, 819-21.
Optical hydroxyl radical sensor Sutton, H.C. (1985). Efficiency of chelated iron compounds as catalysts for the Haber-Weiss reaction. J. Free Rad. Biol. Med., 1, 195-202. Walling, C. & Johnson, R.A. (1975). Fenton’s reagent. V: Hydroxylation and side-chain cleavage of aromatics. J. Am. Chem. Sot., 97, 363-7.
329