A Retrograde Double-Labeling Technique for Light Microscopy a Combination of Axonal Transport of Cholera Toxin B-Subunit and a Gold-Lectin Conjugate T.J.H
Total Page:16
File Type:pdf, Size:1020Kb
WETNODS Journal of Neuroscience Methods 61(1995) 127-138 A retrograde double-labeling technique for light microscopy A combination of axonal transport of cholera toxin B-subunit and a gold-lectin conjugate T.J.H. Ruigrok *, T.M. Teune, J. van der Burg, H. Sabel-Goedknegt Department of Anatomy, Erasmus University, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands Received 5 September 1994; revised 23 February 1995; accepted 24 February 1995 Abstract A light microscopical,non-fluorescent, retrograde double-labelingtechnique is described.Cholera toxin B-subunit (Cl%) and a conjugateof wheatgermagglutinin and bovine serumalbumin coupled to 10 nm gold particles (gold-lectin) are both excellent retrograde tracers and, when visualized by meansof immunohistochemistryand silver intensification, respectively, may be readily identified within the samecell. This light microscopicalretrograde double-labelingtechnique is illustrated in rat with experimentsdesigned to investigate the collateralisation(1) of vestibular neurons to the spinal cord and oculomotor complex, (2) of spinal neuronsto the left and right lateral reticular nucleus, and (3) of inferior olivary neuronsto the uvula of the cerebellum. Advantages over fluorescent double-labelingexperiments are found in the fact that the diaminobenzidinereaction product as well as the silver/gold depositsdo not fade and can be examinedin counterstainedsections. Moreover, the injection sitescan be kept quite small and may be guided by electrophysiologicalrecording through the injection pipette. Keywords: Retrograde tracing; Fluorescent tracer; Axon collateral; Vestibular nuclei; Lateral reticular nucleus; Cerebellum; Inferior olive 1. Introduction material is usually arduous since it has to be performed with the aid of fluorescence microscopy in which cy- The introduction of the fluorescent retrograde dou- toarchitectural boundaries are difficult to identify. ble-labeling technique by Kuypers and collaborators When tracers are used with different excitation wave- (Bentivoglio et al., 1980a,b; Huisman et al., 1983; Keizer lengths, filters may have to be changed. Moreover, et al., 1983; Kuypers et al., 1977, 1980) has aided fluorescence is prone to fading, especially after pro- neuroanatomical research enormously since it became longed storage or illumination. In addition, fluorescent possible to study the collateralisation of axons. More tracers are known to be rather toxic to both the animal recently, the number of fluorescent tracers to choose (they may result in considerable tissue necrosis) (see from has increased dramatically and made triple or e.g.: Keizer et al., 1983) arid the environment, and are, multiple retrograde labeling possible. Combinations of with some exceptions, difficult to dissolve. Because of fluorescent tracers with wheatgerm agglutinated- this last property, injections with these fluorescent horseradish peroxidase (WGA-HRP) have been em- tracers (such as Fast Blue, Diamino Yellow and Nu- ployed also (Wigston and Kennedy, 1987; Kitao et al., clear Yellow), frequently result in large injection sites. 1989). However, there are a number of disadvantages Finally, it is usually not possible to record neuronal in employing fluorescent tracers. The analysis of the activity through the injection pipette to help establish- ing the optimal injection site. For these reasons we searched for a combination of * Corresponding author: Tel.: (31) 10-408-7296; Fax: (31) 10-436- two retrogradely transported tracers that: (1) are both 5780. easily detected with the light microscope; (2) are read- 0165-0270/95/$09..50 8 1995 Elsevier Science B.V. All rights reserved SSDI 0165-0270(94)00034-8 128 T.J.H. Ruigrok et ul. /Journul of Neuroscience Methods 61 (1995) 127-136 ily differentiated from one another even when present made use of the low-salt variety of CTb (List Biological in the same neuron; (3) can be combined with counter- Lab Product #104, lot #CVXG-15A), there was no staining; (4) can be deposited in small-sized injection need to desalt and exchange buffers. Thus, 0.5 mg of sites; (5) allow recording of neuronal activity through lyophilized CTb was dissolved in 50 ~1 of distilled the injection pipettes. water to a final concentration of 1% CIb in 0.2 M We report here that the combination of cholera sodium phosphate (pH 7.51, which were aliquotted in 4 toxin B-subunit (CTb) and a goldsol coupled to the ~1 quantities and stored in the freezer. lectin WGA that was conjugated to bovine serum albu- Glass micropipettes containing an inner filament mine (BSA) fulfills these demands. The use of CI’b as (outer diameter: 2 mm; Clark Electromedical Instru- a retrograde tracer was introduced by Luppi and et al. ments) with tips broken at 12-15 pm were backfilled (1990) and gold-lectin conjugates have first been used with the aliquotted CTb solution using a Hamilton by Menetrey (1985). syringe. The micropipette was connected to a 1.0 ml syringe with a Luer Lock adapter for catheders (Un- imed, Geneva, Switzerland) and pressure was applied 2. Methods manually to the back of the micropipette to the point were CTb solution oozed out of the pipette tip. In this 2.1. Preparation and injection of tracers way, air bubbles trapped in the shank of the pipette were removed. A silver wire, insulated except for the 2.1.1. Cholera toxin B-subunit tip, connected the solution with a Grass preamplifier. Retrograde tracing with CTb results in the visualisa- Electrophysiological recordings were conventionally tion of the soma and a large part of the dendritic tree monitored and stored on tape (Fig. 1). After determi- of the labeled neurons (Luppi et al., 1990; Shinonaga nation of the injection location, the pipette was con- et al., 1992). In addition, CTb is known to be trans- nected to an iontophoresis device, capable of deliver- ported in an anterograde fashion. We used ion- ing constant positive current pulses (7 s on, 7 s off) of 4 tophoretic injections of CTb as originally published by PA for a period of 15-30 min. At the end of this Luppi et al. (1990). However, since we predominantly period, the pipettes were left in situ for 5-10 min to Fig. 1. Schematic diagram of the experimental set-up. Note that electrical signals recorded at the pipette tip can be monitored after which ejection of the tracer may take place by either iontophoresis or pressure injection. T.J.H. Ruigrok et ai./Joumal of Neuroscience Methods 61 (1995) 127-138 129 minimize leakage of the tracer along the pipette track. months at 4°C without losing its retrograde transport In some types of experiments CTb was pressure-in- capabilities. jected (e.g., in the spinal cord). Gold-lectin conjugates are not suited to be deliv- ered iontophoretically. Therefore, in order to be able 2.1.2. Goldsol coupled to WGA-BSA conjugate to make restricted injections a pressure device was Gold-lectin conjugates have been introduced as a used (Gibson et al., 1987). A small amount (about 1 ~1) neuroanatomical tracer by Menetrey and colleagues of the WGA-BSA-gold solution was sucked into a (Basbaum and MenCtrey, 1985; Menetrey and Lee, glass micropipette (tip diameter: 12-15 pm) by apply- 1985; Menetrey, 1987) and have been demonstrated to ing vacuum to the back of the pipette, which was be transported in a strictly retrograde direction Luer-locked to the pressure delivering device (Fig. 1). (Llewellyn-Smith et al., 1992). Initially, goldsol was We empirically established that electrophysiological coupled to WGA-HRP or WGA-apoHRP. Here, we recordings through the WGA-BSA-gold solution could report that 10 nm goldsol (Aurion) coupled to a conju- be made possible by introducing a thin (diameter: 75 gate of WGA and BSA also acts as an excellent, but pm) and except for the very tip, insulated silver wire much cheaper, retrograde tracer. The following steps into the pipette prior to filling (Fig. 1). After establish- (modified after Roth, 1983) were performed to prepare ing the optimal injection location, pressure pulses of a tracer of a good, reproducible, quality. varying intensity were applied to inject the tracer. The Step 1: preparation of the WGA-BSA conjugate. Dis- total injected volume (in nl) was estimated to be equal solve 1 mg WGA (Sigma) and 4 mg BSA in 0.25 ml of to 1/4rrd2h, where d is the approximate diameter (in 0.005 M NaCl. Adjust pH to 7.0 with either 0.2 M mm) of the gold-lectin meniscus in the shank of the H,PO, or 0.2 M K&O,. Add 50 ~1 of 0.25% glu- injection pipette and h the displacement (in mm) of taraldehyde (E.M. grade) and stir for 2 h at room the meniscus during injection. In this way, volumes temperature. Finally, add 12.2 ml of 0.005 M NaCl and ranging between 10 and 1000 nl WGA-BSA-gold could dialyze overnight in 0.005 M NaCI. Total volume of the be reliably injected. protein solution is 12.5 ml and contains 5 mg of pro- 2.2. Surgical procedure tein. Step 2: preparation of the WGA-BSA-gold conjugate. Injections were made in male Wistar rats (200-250 Prepare the following solutions in clear glass or plastic g) which had been anesthetized with pentobarbital(l20 vials: mg/kg, i.p.) and were mounted in a stereotactic device. Injection sites were roughly determined with the aid of 90 ~1 of protein + 10 ~1 of 0.005 M NaCl; the stereotactic atlas of Paxinos and Watson (1986) but 70 ~1 of protein + 30 ~1 of 0,005 M NaCl; were verified by recording neuronal activity typical of 50 ~1 of protein + 50 ~1 of 0.005 M NaCl; region to be injected through the injection pipette. 30 ,ul of protein + 70 ~1 of 0.005 M NaCl; Target areas were combinations of inferior olivary nu- 10 ~1 of protein + 90 ~1 of 0.005 M NaCl.