Liquid membrane electrodes have greatly facilitated the measurement of pH and ion activities in intra- and extracellular fluids. Ion-selective microelectrodes (ISMEs) can be constructed by introducing a short column of a neutral carrier or ion-exchanger cocktail into the tip of a glass micropipette. The inner surface of the glass is first made hydrophobic by exposure to compounds containing silane (Thomas, 1978; Amman, 1986). This process, termed silanization, prevents subsequent displacement of the hydrophobic cocktail, either by entry of aqueous biological fluids through the tip of the micropipette or by migration of the backfilling solution. Macroscopic electrodes (5–10 mm diameter) can be produced by first incorporating appropriate neutral carriers into solvent polymeric membranes of polyvinyl chloride (PVC) or silicone rubber. Membranes can then be stamped out with a cork borer and glued or fused onto PVC tubing or commercially available electrode bodies (Meier et al. 1980). Electrodes with outer diameters of 1.5–1.7 mm have been formed by dipping tubing into a cocktail containing the ionophore, PVC and solvents. Solvent evaporation forms a membrane about 25 μm thick on the tip of the tubing (Oesch et al. 1987). A technique for the construction of miniature double-barrelled K+ electrodes with a total o.d. of 600 μm and a shank 1.3 cm long has been described by Hill et al. (1978).

Although the latter type of electrode is suitable for intramyocardial and intravascular recording from animals the size of dogs and pigs (Hill et al. 1978), flexible electrodes with smaller diameters and longer shanks are required for recording ion activity in the circulatory system or body cavities of smaller animals such as insects. This paper describes techniques for the simple and rapid construction of flexible pH- and ion-selective electrodes with external diameters as small as 40μm. Double-barrelled electrodes with total outer diameters of 80–100 μm and shanks as long as 50 cm can be inserted into the body cavity of insects such as locusts through syringe needles as fine as 26 gauge.

The electrode body is made from a disposable polyethylene 1ml syringe (Becton-Dickinson & Co., Rutherford, NJ). A 1–2cm region of the barrel is heated over a low flame until it softens and the diameter increases slightly (Fig. 1A). The syringe is then drawn out into a fine tube about 0.1 mm in diameter (Fig. 1B). The procedure to this point is that described previously for producing a disposable syringe with integral plastic needle for use in backfilling glass micropipettes (see Fig. 10 of Thomas, 1978). During or after cooling, the fine tube can be pulled out further by hand to a diameter as small as 40–50 /tm and cut to length with a razor blade. The barrel is then cut to within about 1 cm of the top of the shank. About 0.2ml of an appropriate internal reference solution is injected into the back of the syringe, and forced to the tip of the tubing by the application of pressure either by mouth or by re-inserting the plunger and pressing lightly (Fig. 1C). Under a dissecting microscope, the tip is then inserted into a fine glass capillary tube (Fig. 1D) containing about 1 μl of the appropriate ionophore cocktail (Table 1). The cocktail can be used as supplied if the electrodes are to be used within 1 week. Modifications to provide electrodes with longer lifetimes are described below. After a column of about 1–2 mm (approximately 20 nl) of ionophore cocktail has entered the tip by capillarity, the tubing is retracted from the glass tube. Silanization is not required because the syringe tubing is hydrophobic. Electrodes can be maintained in physiological saline or internal reference solution until use. With practice, the time of fabrication and filling can be reduced to less than 3 min per electrode.

Table 1.

Selectivity factors (logKjj for flexible, subminiature ion-selective electrodeas

Selectivity factors (logKjj for flexible, subminiature ion-selective electrodeas
Selectivity factors (logKjj for flexible, subminiature ion-selective electrodeas
Fig. 1.

Steps in the construction of a flexible subminiature ISE. Further details in text. (A) A disposable plastic syringe is rotated (arrows) as it is heated over a low flame.(B) The heated region bulges slightly and becomes more translucent when the plastic is near its melting point. For double-barrelled pipettes, the melted regions of two identical syringes are pushed together so that they fuse. Single or fused double syringes are then drawn out (arrow) by holding vertically and pulling in the direction of the arrow. (C) Internal filling solution is forced to the tip of the fine tubing, which is of nearly constant diameter over lengths up to 50 cm. (D) The tip of the tubing is placed into a glass capillary tube containing ionophore cocktail, and allowed to fill by capillarity until a 1–2 mm long column is present. A chloride-coated silver wire is then introduced into the back of the internal reference solution.

Fig. 1.

Steps in the construction of a flexible subminiature ISE. Further details in text. (A) A disposable plastic syringe is rotated (arrows) as it is heated over a low flame.(B) The heated region bulges slightly and becomes more translucent when the plastic is near its melting point. For double-barrelled pipettes, the melted regions of two identical syringes are pushed together so that they fuse. Single or fused double syringes are then drawn out (arrow) by holding vertically and pulling in the direction of the arrow. (C) Internal filling solution is forced to the tip of the fine tubing, which is of nearly constant diameter over lengths up to 50 cm. (D) The tip of the tubing is placed into a glass capillary tube containing ionophore cocktail, and allowed to fill by capillarity until a 1–2 mm long column is present. A chloride-coated silver wire is then introduced into the back of the internal reference solution.

The following ionophore cocktails were obtained from Fluka Chemical Corp., Ronkonkoma, NY: (1) Ca2+ ionophore I (ETH 1001), cocktail B; (2) Ca2+ ionophore II (ETH 129), cocktail A; (3) H+ ionophore I (tridodecyl amine, TDDA), cocktail B; (4) H+ ionophore II (ETH 1907), cocktail A; (5) K+ ionophore I (valinomycin), cocktail A; (6) Na+ ionophore II (ETH 157), cocktail A. For NH4+ electrodes, a cocktail was made by mixing 10 % NH4+ ionophore I (75 % nonactin, 25 % monactin) with 89 % 2-nitrophenyl octyl ether and 1 % sodium tetraphenylborate. Internal reference solutions for each ion (in parentheses) were as follows: (Na+), 0.5 mol 1-1 NaCl; (K+), 0.5 mol 1-1 KC1; (H+), 0.1 mol 1-1 NaCl and 0.1 mol 1-1 sodium citrate, adjusted to pH 6 (Thomas, 1978); (Ca2+), 0.2 mol 1-1 KC1, 5 mmol 1-1 CaC12 and 10 mmol 1-1 EGTA, yielding pCa 7 (Alvarez-Leefmans et al. 1981).

Recording from the subminiature ISE was accomplished by placing silver chloride-coated silver wires into the internal reference solution and connecting the wires to an operational amplifier. Noise due to electrical interference was reduced by filtering the amplifier output through a low-pass RC filter with a time constant of Is. Electrode response times were of the order of 2–4s. The resistance of the subminiature ISEs varied from Ix108-5x108Ω, comparable to or less than the 109Ω typical for solvent polymeric electrodes and the 109-1011 Ω of ISMEs. The potentials generated by the subminiature ISEs, therefore, can accurately be measured by operational amplifiers with input impedances of 1011 Ω or more. These are typically used for recording from standard glass microelectrodes filled with 3 mol 1-1 KC1, and are cheaper and more readily available than the high-impedance (>1014Ω) operational amplifiers typically used for ISMEs.

Ionophore cocktail tended to leak out of the tip of the electrode if use involved inadvertent contact between the tip and a hydrophobic surface such as an acrylic or polyvinyl chloride experimental chamber. This problem was eliminated by filling the tip with a mixture of one part of an ionophore cocktail with two parts of 15 % (w/v) polyvinyl chloride in tetrahydrofuran (Fluka). Evaporation of the solvent resulted in a tough gel-like membrane that was not easily displaced or dislodged by contact of the tip with hydrophobic surfaces. Additional advantages were that the lifetime of electrodes stabilized with PVC exceeded 4 weeks, and that hydrostatic pressures within very long (>50cm) electrodes did not force out the sensor, as sometimes occurred in the absence of PVC. Electrode resistance and response times were not noticeably altered by the addition of PVC.

For reference electrodes, plastic syringes were heated and pulled as for ISEs and were backfilled either with 3 mol 1”1 KC1 (for pH, Na+, Ca2+) or with NaCl (for K+ measurements) in 3% agar. Double-barrelled subminiature ISEs were fabricated by heating two plastic syringes, as described above, then joining the barrels while the plastic was near its melting point. The fused barrels could then be pulled out to the desired total diameter (80–100μm), as for single-barrelled ISEs, and one barrel used for a reference while the other was used for measurement of ion activity.

Typical slopes per decade change in ion activity were 56–58 mV for H+, 52–55 mV for Na+ and K+, 50 mV for NH4+ and 27–29 mV for Ca2+. Selectivity of an electrode for an interfering ion relative to the primary ion sensed by the ionophore is described by selectivity factors (Kij), usually given as log A7j, where ‘i’ refers to the primary ion and ‘j’ to the interfering ion. A value of —2, for example, indicates that the electrode is 100 times more selective for the primary ion, i, than for the interfering ion, j. Measured selectivity factors of subminiature ISEs are summarized in Table 1, together with the calculated selectivity factors required for a maximal interference of 1% (Kij.max)-The latter were calculated for a representative extracellular fluid containing 135–150 mmol 1-1 Na+, 3.5–10 mmol 1-1 K+, 0.45–2 mmol 1-1 ionized Mg2+ and 1–2 mmol 1-1 ionized Ca2+, at a pH of 6.5–7.5. Required selectivity factors were calculated from the equation of Oesch et al. (1986):
where Kij max is the highest tolerable value of the selectivity factor, ai,min is the lowest expected activity of the primary ion i, aj,max is the highest expected activity of the interfering ion j, pj is the highest tolerable error in measured activity of i due to interference by j, and Zi, Zj are the valencies of i and j, respectively. Providing that logKi j is more negative than logKij max, the selectivity is sufficient to measure the primary ion with less than 1 % interference. Table 1 shows that with currently available ionophore cocktails, subminiature ISEs are suitable for measurement of H+, Ca2+ and K+ in typical extracellular fluids. For 135mmoll-1 Na+, 10mmol 1-1 K+ results in 2.5 % interference. However, in fluids in which the ratio aK/aNa is less than 0.03, the interference of K+ on the response of subminiature Na+-selective electrodes is less than 1 %.

Flexible subminiature ISEs may be suitable for recording near or on mucosal surfaces, such as gills, bronchial passages or the oesophagus. Levels of pH and NH4+ in the micro-environment near the gills of trout, for example, are different from those in the bulk water (Playle and Wood, 1989), and such local ion gradients might be detectable using flexible subminiature ISEs. An example of measurement of the surface pH of moistened filter paper is shown in Fig. 2A. The flexible double-barrelled electrode was advanced until slight bending of the tubing indicated contact with the surface of the filter paper. This type of ISE may also be useful for the measurement of bathing fluid pH or ion activity in small experimental chambers used for in vitro perfusion, particularly if holes or passageways, which need not be straight, are milled in the chamber. They can also be used for analysis of nanolitre droplets of biological fluids collected in micropipettes and expelled under paraffin oil.

Fig. 2.

Representative recordings obtained with flexible, double-barrelled subminiature pH electrodes based on the ionophore ETH 1907. A different electrode was used for each recording. (A) The electrode tip was pressed against filter paper (FP) moistened with insect saline (O’Donnell, 1985) adjusted to pH 6.5, then transferred through the air to a second piece moistened with saline at pH7.5. The recording went off-scale during the transfer (through air) between filter papers. For comparison, recordings when the electrode was dipped into baths (Petri dishes) filled with the same salines (bath pH6.5 and 7.5) are shown. (B) Recording of haemolymph pH from an adult male locust, Locusta migratoria. The electrode was inserted through a 26 gauge needle, which pierced the arthrodial membranes of the ventral surface of the abdomen. The electrode was calibrated in pH6.5 and 7.5 solutions before and pH7.5 solution after haemolymph pH measurement. (C) Recording of crop pH of an adult male locust. The digestive tract was dissected and pinned out in a dish of saline. The double-barrelled pH electrode was transferred through the air from a dish of pH 7.5 saline and introduced into the crop through a 2 cm length of fire-polished glass pipette (1.9 mm o.d., 0.6mm i.d.), 2–3 mm of which was pushed into the cut end of the oesophagus. The pipette was then withdrawn. Small changes in crop pH were coincident with muscular contractions and resulting movements of crop contents. After withdrawal from the crop, the electrode was calibrated in saline at pH6.5 and 7.5.

Fig. 2.

Representative recordings obtained with flexible, double-barrelled subminiature pH electrodes based on the ionophore ETH 1907. A different electrode was used for each recording. (A) The electrode tip was pressed against filter paper (FP) moistened with insect saline (O’Donnell, 1985) adjusted to pH 6.5, then transferred through the air to a second piece moistened with saline at pH7.5. The recording went off-scale during the transfer (through air) between filter papers. For comparison, recordings when the electrode was dipped into baths (Petri dishes) filled with the same salines (bath pH6.5 and 7.5) are shown. (B) Recording of haemolymph pH from an adult male locust, Locusta migratoria. The electrode was inserted through a 26 gauge needle, which pierced the arthrodial membranes of the ventral surface of the abdomen. The electrode was calibrated in pH6.5 and 7.5 solutions before and pH7.5 solution after haemolymph pH measurement. (C) Recording of crop pH of an adult male locust. The digestive tract was dissected and pinned out in a dish of saline. The double-barrelled pH electrode was transferred through the air from a dish of pH 7.5 saline and introduced into the crop through a 2 cm length of fire-polished glass pipette (1.9 mm o.d., 0.6mm i.d.), 2–3 mm of which was pushed into the cut end of the oesophagus. The pipette was then withdrawn. Small changes in crop pH were coincident with muscular contractions and resulting movements of crop contents. After withdrawal from the crop, the electrode was calibrated in saline at pH6.5 and 7.5.

A major advantage of both single- and double-barrelled subminiature ISEs is that they can be inserted into a blood vessel or other tissue compartment through syringe needles as small as 26 gauge (Fig. 2B) or through a fire-polished glass capillary tube (Fig. 2C). The syringe needle or glass capillary can be withdrawn after the subminiature ISE and reference electrode are in place. The flexibility of the subminiature ISE not only prevents damage to the electrode itself but also limits damage to the preparation, for example to the gut wall during peristalsis.

Although suitable for haemolymph pH measurements in insects over short periods (5–10min), longer-term use resulted in blockage of the ISE tip by haemocyte clotting. Electrodes blocked in this way could be restored by simply cutting off about 0.2 mm of the tip with a razor blade, thereby exposing a fresh surface of the ionophore cocktail.

This work was supported by NSERC operating grants. The author is grateful to C. M. Wood for his comments on a draft of the manuscript.

Alvarez-Leefmans
,
F. J.
,
Rink
,
T. J.
and
Tsien
,
R. Y.
(
1981
).
Free calcium ions in neurons of Helix aspersa measured with ion-selective microelectrodes
.
J. Physiol., Lond
.
315
,
531
538
.
Ammann
,
D.
(
1986
).
Ion-Selective Microelectrodes. Principles, Design and Application
.
Berlin
:
Springer-Verlag
.
Ammann
,
D.
and
Anker
,
P.
(
1985
).
Neutral carrier sodium ion-selective microelectrode for extracellular studies
.
Neurosci. Lett
.
57
,
287
.
Ammann
,
D.
,
Lanter
,
F.
,
Steiner
,
R. A.
,
Schulthess
,
P.
,
Shiuo
,
Y.
and
Simon
,
W.
(
1981
).
Neutral carrier based hydrogen ion selective microelectrode for extra- and intracellular studies
.
Analys. Chem
.
53
,
2267
2269
.
Ammann
,
D.
,
Morf
,
W. E.
,
Anker
,
P.
,
Meier
,
P. C.
,
Pretsch
,
E.
and
Simon
,
W.
(
1983
).
Neutral carrier based ion-selective electrodes
.
Ion-Selective Electrode Rev
.
5
,
3
92
.
Buhrer
,
T.
,
Peter
,
H.
and
Simon
,
W.
(
1988
).
NH4+ ion-selective microelectrode based on the antibiotics nonactin/monactin
.
Pflügers Arch
.
412
,
359
362
.
Chao
,
P.
,
Ammann
,
D.
,
Oesch
,
U.
,
Simon
,
W.
and
Lang
,
F.
(
1988
).
Extra- and intracellular hydrogen ion-selective microelectrode based on neutral carriers with extended pH response range in acid media
.
Pflügers Arch
.
411
,
216
219
.
Hill
,
J. L.
,
Gettes
,
L. S.
,
Lynch
,
M. R.
and
Hebert
,
N. C.
(
1978
).
Flexible valinomycin electrodes for on-line determination of intravascular and myocardial K+
.
Am. J. Physiol
.
235
,
H455
H459
.
Lanter
,
F.
,
Steiner
,
R. A.
,
Ammann
,
D.
and
Simon
,
W.
(
1982
).
Critical evaluation of the applicability of neutral carrier-based calcium selective microelectrodes
.
Analys. chim. Acta
135
,
51
59
.
Meier
,
P. C.
,
Ammann
,
D.
,
Morf
,
W. E.
and
Simon
,
W.
(
1980
).
Liquid-membrane ion-selective electrodes and their biomedical applications
.
In Medical and Biomedical Applications of Electrochemical Devices
(ed.
J.
Koryta
).
Chichester
:
J. Wiley and Sons
.
O’donnell
,
M. J.
(
1985
).
Calcium action potentials in the developing oocytes of an insect, Rhodnius prolixus
.
J. exp. Biol
.
119
,
237
300
.
Oehme
,
M.
and
Simon
,
W.
(
1976
).
Microelectrode for potassium ions based on a neutral carrier and comparison of its characteristics with a cation exchanger sensor
.
Analys. chim. Acta
86
,
21
25
.
Oesch
,
U.
,
Ammann
,
D.
and
Simon
,
W.
(
1986
).
Ion-selective membrane electrodes for clinical use
.
Clin. Chem
.
32
,
1448
1459
.
Oesch
,
U.
,
Brzozka
,
Z.
,
Xu
,
A.
and
Simon
,
W.
(
1987
).
Solvent polymeric membrane pH catheter electrode for intraluminal measurements in the upper gastrointestinal tract
.
Med. Biol. Eng. Comput
.
25
,
414
419
.
Playle
,
R. C.
and
Wood
,
C. M.
(
1989
).
Water chemistry changes in the gill micro-environment of rainbow trout: experimental observations and theory
.
J. comp. Physiol. B
159
,
527
537
.
Schefer
,
U.
,
Ammann
,
D.
,
Pretsch
,
E.
,
Oesch
,
LJ.
and
Simon
,
W.
(
1986
).
Neutral carrier based Ca2+-selective electrode with detection limit in the subnanomolar range
.
Analys. Chem
.
58
,
2282
2285
.
Thomas
,
R. C.
(
1978
).
Ion-Sensitive Intracellular Microelectrodes. How to Make and Use Them
.
New York
:
Academic Press
.
Walker
,
J. L.
(
1971
).
Ion-specific liquid ion exchanger microelectrodes
.
Analys. Chem
.
43
,
89A
93A
.