(B) Retrograde labeling
(1) Influence of the immunohistochemical protocol
In an attempt to maximize the CTb retrograde labeling, we tested the sensitivity of different immunohistochemical protocols and the influence of several parameters such as the incubation duration and temperature for each step of these procedures.
In a first series of experiments, we compared the sensitivity of two different secondary antibodies and that of the goat PAP, the ABC-HRP complex and streptavidin-HRP. For this purpose, adjacent sections were incubated in CTb antiserum at a 1:40,000 dilution for 4 days at 4 °C and then submitted to 4 different protocols:
(a) donkey anti-goat IgG (1:1,000) and goat PAP (1:1,000);
(b) biotinylated swine anti-goat IgG (1:2,000) and ABC-HRP complex (1:500);
(c) biotinylated donkey anti-goat IgG (1:2,000) and ABC-HRP complex (1:500);
(d) biotinylated donkey anti-goat IgG (1:2,000) and streptavidin-HRP complex (1:40,000).
Incubations in each compound lasted 90 min at room temperature in PBST. The sections were then reacted using the standard DAB-nickel procedure (see Materials and Methods).
Using any one of the protocols, the retrogradely labeled cells were characterized by the presence of dark blue granular reaction products in their cytoplasm ofter extending to their proximal dendrites. Nevertheless, we found that, when using the classical PAP method, the labeling was very faint when compared with that observed using the ABC-HRP complex and streptavidinHRP. The best results were obtained with streptavidinHRP associated with the biotinylated donkey anti-goat IgG (protocol d). Indeed, using such a protocol, the number, intensity and dendritic labeling of the retrogradely labeled neurons dramatically increased. We further determined, comparing the results obtained with protocols b, c and d, that this increase in sensitivity was due to the combined superiority of streptavidin-HRP and the biotinylated donkey antigoat IgG over the biotinylated swine anti-goat IgG and the ABC-HRP complex. It is important to note that this dramatic gain in sensitivity was not associated with an increase in the diameter of the injection sites as shown in Fig. 1B and D (see general remarks).
In our first reports using the CTb retrograde tracing method 34 36, we used the ABC-HRP complex and biotinylated swine anti-goat IgG (protocol b) to reveal CTb, while in the recent studies we introduced the more sensitive streptavidin-HRP and biotinylated donkey antigoat IgG (protocol d) 14 15 58-60. Therefore, we found it necessary to quantify the gain of sensitivity obtained when using these last two products as well as their respective contribution to this increase. For this purpose, we counted the number of CTb cells on adjacent sections stained with either b, c or d protocols. We found an 84% increase in the number of retrogradely labeled cells between the b and d protocols (when counting neurons on adjacent sections labeled with protocols b and d). This increase in sensitivity was due for 45% to the biotinylated donkey anti-goat IgG (deducted from adjacent sections labeled with protocols b and c) and for 25% to streptavidin-HRP (deducted from cell count on adjacent sections stained with protocols c and d).
In a second series of experiments, we compared the intensity and aspect of the retrograde labeling according to the length and the temperature of the incubation, in the primary and secondary antibodies as well as in streptavidin-HRP. We were able to reveal the injection sites and the retrograde labeling even after a short overnight incubation at room temperature in CTb antiserum (1:20,000) followed by a 1 h 30 min incubation in the linking antiserum (1:2,000) and streptavidin-HRP (1 :40,000). This protocol allowed us to examine the results 24 h after cutting the sections. However, this fast procedure gave less intense and less complete retrograde labeling than the slow procedure consisting of a 3-4 day incubation at 4 °C in CTb antiserum (1:40,000) and 90 min incubations at room temperature in the linking antibody (1:2000) and streptavidin-HRP (1:40,000). In such conditions, the staining of the retrogradely labeled cells was intense and granular and the background staining was very limited. Therefore, this immunohistochemical procedure was ideally suited for double labeling experiments (see below). When using the same concentration of linking antibody and streptavidin-HRP, but with an overnight incubation at 4 °C, the intensity of the retrograde labeling further increased and in some cases the cell bodies were completely filled with black granules. However, we also observed a concomitant increase of the background.
On the other hand, we determined that the CTb immunostaining did not significantly decrease on sections stocked in PBST-AZ or mounted on slides. We also found a better preservation of the morphology and a more complete staining of the retrogradely labeled cells on vibratome than on cryostat sections. Indeed, as illustrated in Fig. 4, the CTb retrograde labeling was Golgi-like with an extensive filling of the dendritic trees.
(2) Influence of the fixative
We checked several types of fixative with variable concentrations of PF, GLU, carbodiimide and PA. We obtained the best results using a fixative composed of 4% PF, 0.1% GLU and 0.2% PA. When using an increasing concentration from 0.1 to 5% of GLU, we observed a concomitant gradual decrease in intensity of the CTb labeling and an increase of the background staining. With 5%, we were unable to distinguish the specific staining from the background. However, this phenomenon can be almost eliminated and the specific stain reappeared when the sections were pretreated with 0.4% borohydride in PBS for 5-15 min.
The addition of as little as 0.2% carbodiimide to a 4% PF fixative gave rise to unexpected artefactual labeling of the myelinated fibers and a less sharp although intense punctate retrograde labeling.
(3) Characteristics of CTb as a retrograde tracer
After CTb pressure and iontophoretic injections, even after the shortest survival (24 h) and for the most limited iontophoretic applications (200 µm in diameter), retrograde labeling was observed in all investigated pathways and there was no absence of uptake or transport in any particular neuronal groups. These findings indicate that, in contrast with nearly all the other retrograde tracers, CTb can be used to map the afferents to very small groups of cells. Large iontophoretic injections sites (400-800 um) gave rise to an extensive retrograde labeling comparable with the smaller pressure injections (0.05 µl, 600-800 µm in diameter).
No evidence of anterograde and retrograde transsynaptic labeling was observed even after 18 days of survival.
We also found no decrease in the number of retrogradely labeled cells nor in the intensity of their staining in cats with 72 h of survival including 24 or 48 h of colchicine treatment nor in those with 10, 15 and 18 days of survival. On the contrary, as illustrated in Fig. 3C and D, we observed an increase in the staining of the retrogradely labeled neurons as well as a much more extensive labeling of their dendritic trees after 10, 15 and 18 days of survival.