- Email: [email protected]
Wheat germ agglutinin conjugated to TRITC: A novel approach for labeling primary projection neurons of peripheral afferent nerves
Journal of Neuroscience Methods 93 (1999) 139 – 147 www.elsevier.com/locate/jneumeth
Wheat germ agglutinin conjugated to TRITC: A novel approach for labeling primary projection neurons of peripheral afferent nerves Andrea Sawczuk a,c,*, David A. Covell, Jr. b a
Departments of Oral Pathology, Biology, and Diagnostic Sciences, Uni6ersity of Medicine and Dentistry of New Jersey, 1110 Bergen Street, Newark, NJ 07103, USA b Department of Orthodontics, Uni6ersity of Medicine and Dentistry of New Jersey, 1110 Bergen Street, Newark, NJ 07103, USA c Department of Neuroscience, Uni6ersity of Medicine and Dentistry of New Jersey, 1110 Bergen Street, Newark, NJ 07103, USA Received 11 March 1999; received in revised form 12 August 1999; accepted 20 August 1999
Abstract Wheat germ agglutinin conjugated to tetramethylrhodamine isothiocyanate-dextran (WGA – TRITC) was studied as a novel tracer of primary projection neurons of pharyngeal (PhN) and superior laryngeal (SLN) branches of the vagus nerve. The SLN and PhN were dissected from rat cervical tissues and the proximal end of the nerves were bathed in tracer for 60 – 90 min. The animals were sacrificed 42–72 h later. The tissue was fixed, sliced, mounted on slides and viewed under epifluorescence. The clarity of the fluorescent label in projection neurons was confounded in some regions of the brainstem by autofluorescence. A computer image analysis method was developed to quantify fluorescence intensity for definitive identification of labeled neurons. Brainstem neurons labeled by afferent projections of the SLN and PhN were localized to the nucleus tractus solitarius. Efferents were identified in the nucleus ambiguus. WGA–TRITC labeled cells were observed in the ipsilateral brainstem at intensities significantly different from the fluorescence observed in controls (PB 0.01). The distribution and density of labeling is in agreement with results of previous investigations, suggesting that WGA – TRITC is a useful alternative for tracing SLN and PhN projections to brainstem nuclei. © 1999 Elsevier Science B.V. All rights reserved. Keywords: WGA – TRITC; Fluorescent tracers; Lectin conjugates; Autofluorescence; Vagus nerve; Superior laryngeal nerve; Pharyngeal nerve; Nucleus tractus solitarius; Nucleus ambiguus
1. Introduction Wheat germ agglutinin conjugated to horse-radish peroxidase (WGA– HRP) is the preferred tracer of peripheral nerve central projections and pathways (Altschuler et al., 1989; Mrini and Jean, 1995; Furusawa et al., 1996) because the WGA lectin facilitates transynaptic movement of the HRP label into brainstem neurons (Broadwell and Balin, 1985; Beiger and Hopkins, 1987). Recently, WGA conjugated to the fluorescent tracer, tetramethylrhodamine isothiocyanate-dextran (WGA – TRITC) has become commercially available. Although WGA – TRITC has been * Corresponding author. Tel.: +1-973-972-2454; fax: + 1-973-9723164. E-mail address: [email protected] (A. Sawczuk)
routinely applied to non-neuronal cell biology investigations (Alonso-Varona et al., 1995; Rev. in Haugland, 1996; Menghi et al., 1997), this tracer has had limited use in neuroscience (Naito et al., 1996) and has not previously been used to trace peripheral nerve brainstem projections. Fluorescent tracers have proven to be powerful tools for identifying neuronal pathways (Bentivoglio and Chen, 1993; Horner and Kummel, 1993; Richmond et al., 1994; Schmued, 1994) and their many applications could be further expanded to include transynaptic movement if used as a conjugate with WGA. We desired the combined lectin-fluorophore properties of WGA–TRITC to identify primary brainstem projection neurons of superior laryngeal (SLN) and pharyngeal branches of the vagus nerve (PhN). Afferents of the SLN and PhN are known to project via the nodose ganglion to neurons of the nucleus tractus
0165-0270/99/$ - see front matter © 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 0 2 7 0 ( 9 9 ) 0 0 1 4 1 - 7
140
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
solitarius (NTS; Mrini and Jean, 1995; Furusawa et al., 1996). Efferents project from motoneurons in the caudal nucleus ambiguus (NA) to the muscles of the pharynx and larynx (Medda et al., 1994). Although these nerves provide peripheral neural control for swallowing and respiration, their central pathways segregate functionally in the NTS. This functional segregation suggests that respiration and swallowing can be studied independently once the projection neurons have been identified and characterized (Dobbins and Feldman, 1995). The primary purpose of this study was to determine if WGA–TRITC could be used to identify SLN and PhN brainstem projection neurons and could provide sufficient morphological detail of the labeled neurons to be useful as a vital stain. To accomplish this goal, we required a fluorescent label that was sufficiently robust to be discriminated from brainstem autofluorescence. A method for image analysis was developed to quantitatively distinguish labeled neurons from background autofluorescence and to measure the size of labeled neurons.
2. Materials and methods
2.1. Tracer application Adult Sprague–Dawley rats (Taconic; n = 18) weighing 200–450 g were anesthetized by an intraperitoneal injection of ketamine and xylazine (80 and 12 mg/kg, respectively). A midline incision was made from the mid-trachea to the mandible under aseptic conditions. All remaining dissection was performed on the right side only. The submandibular glands and muscles overlying the trachea were gently moved aside to expose the trachea at the level of the larynx and hyoid bone. The posterior belly of the digastric muscle was detached from the hyoid bone and the hyoid bone was lifted to expose the carotid sheath. The superior laryngeal and pharyngeal nerve branches of the vagus were identified beneath the hyoid bone and 1 cm segments of nerve were dissected from the underlying tissue and the carotid sheath. The nerves were cut at their most distal unbranched aspects and the proximal ends were crushed and isolated with a paraffin cuff that was secured to the proximal nerves with a droplet of cyanoacrylate glue. The cut nerve ends were bathed in 3 – 5 ml of 5% WGA – TRITC in saline (Sigma). Salinesoaked cotton was placed over the nerves and exposed muscles during this time to prevent dehydration. After 60 – 90 min, the remaining tracer was carefully rinsed from the nerves, the paraffin with the adherent cyanoacrylate was removed, the surrounding tissue was inspected for tracer leakage, and the surgical site was lavaged with saline. Since the tracer solution was pink
in color, it was readily identified in the surrounding tissue. Data were not collected from any animal with suspected tracer leakage. Following saline irrigation, the nerves, muscles, and glands were replaced in their original positions and the incision was closed with wound clips. Control animals (n = 6) were divided equally into 2 groups. One group received sham surgeries in which the procedures described above were performed but the tracer was not applied. Animals in the other control group were perfused but no surgeries were performed.
2.2. Perfusion and histology Because WGA–TRITC had not been previously applied to peripheral nerves, we estimated minimum survival times from previous studies with WGA–HRP to be about 48 h. Thus, our survival times ranged between 42 and72 h. (42 h: n=1; 48 h: n=2; 54 h: n = 1; 62 h: n= 2; 66 h: n= 1; 72 h: n= 1). Animals were sacrificed with an overdose of sodium pentobarbital (\ 100 mg/ kg) followed immediately by cardiac perfusion with 50 ml of sodium phosphate buffer and then by 150–200 ml of 4% paraformaldehyde in 0.15 M phosphate buffer at pH 7.2. The brainstems were removed and stored in 4% paraformaldehyde in 0.15 M phosphate buffer with 20% sucrose for about 12 h when they were transferred to 0.15 M phosphate buffer (pH 7.4) with 30% sucrose for an additional 12 h. The brainstems were frozen in a cryostat microtome and sectioned into 40 mm slices. The slices were washed 3–5 times in 0.15 M phosphate buffer at pH 7.4 and placed on slides to dry overnight in the dark. The slides were then rinsed briefly in xylene (B 2.5 s) and coverslipped with a nonfluorescent medium (DPX mountant, Fluka). All animal use was in accordance with the PHS manual ‘Guide to the Care and Use of Laboratory Animals’ and was approved by the Institutional Animal Care and Use Committee of the University of Medicine and Dentistry of New Jersey.
2.3. Data analysis Sections were viewed under a compound light microscope (Olympus BX50WI) using a 40× objective with epifluorescence (excitation wavelength of 525–560 nm). The images were captured with a video camera (DageMTI CCD-300-RC) and digitized with a frame grabber board (LG-3 NuBus, Scion Corporation, Frederick, MD) at consistent microscope, camera, and computer settings. The final images were an integrated average of 16 captured images. Fluorescence intensity was calculated using Scion Image computer software (modified from NIH Image) by applying the following protocol to the unfiltered image: (1) To measure fluorescence intensity of cells
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
within a captured image, the background was subtracted from the signal by manually thresholding the inverted image so that only the punctate areas representing fluorescence were counted (i.e. as pixel gray values); (2) Because SLN projections to the NTS were primarily observed as increased fluorescence of processes interspersed with cells in the interstitial nucleus,
141
measurements of the processes were made within a circle 15 mm in diameter using the manual thresholding technique described above; (3) To test for reliability, 10% of the manually thresholded data were randomly chosen, remeasured by a second blinded investigator and compared with the originally measured data using a paired t-test. There was no significant difference between investigators; (4) To normalize the data for between image comparisons, a fluorescence intensity ratio (FI) was calculated relating signal density to background density using the following formula: Cell Signal Density FI= (Cell Background Density + General Background Density)
Cell signal and background values were determined by averaging three measurements of the same image. General background was calculated as the mean thresholded value of 3 squares, 15 mm/side, taken from representative background areas of the image but avoiding cells, cell processes, and black holes (Fig. 1). Uptake of tracer was considered successful only if FI was determined with an unpaired t-test to be significantly greater than the autofluorescence of controls. Characteristics selected to estimate label acceptability were cell numbers, somata dimensions, and label distribution. The criteria for acceptability were based on the results of previous studies using WGA–HRP and HRP because fluorescent tracers have not previously been used to study the brainstem projection of cervical vagal nerves. Cell number was counted within a 3.6 mm2 area in those locations demonstrating tracer. Cell size was calculated by averaging the lengths of the major and minor axis of 3 cell somata. Surface area was determined with automated ROI (Region of Interest) analysis features (Scion Image software). Only cells with nuclei were included in the cell count and morphometry. Label distribution was estimated by examining the label pattern in individual neurons and by identifying brainstem projection sites. Fig. 1. Isolation method of neurons and background areas for calculation of fluorescence intensity ratio. A: Image of labeled NTS neurons and processes after a 48 h incubation period (light gray and white areas indicate fluorescent label or autofluorescence). Scale bar= 30 mm. B: The areas to be analyzed in Panel A have been outlined in white: neurons, areas of cell processes (circles), and background (squares). C: The image has been inverted which causes the fluorescent label to appear dark gray or black and the background to appear light gray or white. A neuron of interest has been outlined in white. D: The density slice option in the image analysis program (Scion Image) has been applied and the exposure level has been manually adjusted to a level that preserves the dark gray and black punctate areas (presumed fluorescence label) in the neuron being studied. Other areas in the field having the same intensity are also visible. E: The image threshold has been generated based on the exposure level established in Panel D. The mean integrated density analysis was calculated from the thresholded image for signal and background. Calculation of the fluorescence intensity ratio is discussed in Section 2.3. Scale bar = 15 mm.
3. Results
3.1. Identification of labeled projection neurons Label was clearly located in the NTS, NA, ventral NA, DVM, and HG neurons by 48 h survival time. Fluorescence was discernible in the cell somata and proximal processes (Fig. 2C, D) as pink to red granular inclusions primarily concentrated in the cytoplasm surrounding unstained nuclei and more sparsely distributed in cell processes. Label was also visible as small clumps without distinct nuclei (Fig. 2C) in the solitary tract. These clumps were suggestive of cell processes in cross-section and probably included intra-nuclear neuronal processes and primary afferent nerves.
142
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
Fig. 2. Relative difference in fluorescence intensity can be seen in photomicrographs of NTS (A) and NA (B) from control animals compared to the NTS (C) and NA (D) from labeled animals. The fluorescence intensity ratio for labeled neurons of the NTS was approximately 4 times the fluorescence intensity of control neurons (A vs C) and for labeled neurons of the NA was 3.5 times the ratio of control neurons (B vs D). Scale bar= 30 mm. Table 1 Measurements of Ipsilateral Fluorescence Intensity Ratio (FI)a
Control cell autofluorescence WGA–TRITC labeled cells Control processes autofluorescence WGA–TRITC labeled processes
NTSb (SD)
NAc (SD)
DMVd (SD)
HGe (SD)
0.84 2.53 0.65 2.93
2.35 (0.91) 6.63 (2.89)** – –
1.58 (1.13) 2.34 (1.40)* – –
4.18 (1.35) 9.38 (4.29)** – –
(0.50) (0.90)*** (0.54) (2.23)*
a
Values in parenthesis: Ratio of mean pixel gray values of somata or processes to background. NTS: Nucleus tractus solitarius, c NA: Nucleus ambiguus, d DMV: Dorsal nucleus of the vagus, e HG: Hypoglossal nucleus. * PB0.05, ** PB0.01, *** PB0.001. b
Autofluorescence was also granular and had a similar cytoplasmic distribution to labeled neurons but was yellowish in color (Fig. 2A, B). Although the color was generally different between autofluorescence and label, the most dependable distinguishing feature was fluorescence intensity. The intensity of the WGA– TRITC label was quantitatively verified by the computer image analysis that showed a significant elevation in fluorescence intensity relative to the autofluorescence of controls (P B 0.01). FI of control NTS cells in Fig. 2 was 0.72 and for processes was
1.31, whereas the FI of label in NTS somata (3.33) and processes (7.08) was significantly greater than controls (PB 0.001, Table 1). The FI of control and labeled NA neurons in Fig. 2 was also significantly different (PB 0.01, Table 1). Additionally, tracer was identified in the ipsilateral DMV and HG at intensity levels significantly elevated from controls (Table 1). By 72 h tracer appeared to have moved beyond the primary projection neurons (Fig. 3), labeling the contralateral NTS (FI= 1.68) and NA (FI= 11.00). Tracer was also seen in the contralateral DMV but not in the contralateral HG by 72 h.
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
3.2. Brainstem distribution of labeled projection neurons Fig. 4 shows the distribution of label in cross-sectional drawings from the most caudal to the most rostral aspect of brainstem projections. Fluorescence label was visible in the ipsilateral NTS by 48 h of incubation (* symbols in Fig. 4) from about 3 slices
143
caudal to the obex (120 mm) to 2 mm rostral to the obex. Although fluorescence was most intense in somata and processes of the interstitial subnucleus, rostral areas of the NTS compatible with the intermediate nucleus were also labeled. Fluorescence was most concentrated in the lateral NTS, primarily in processes, and less concentrated in the medial aspect of the nucleus.
Fig. 3. Fluorescence label at 72 h in the ipsilateral NTS (A) and NA (B). Label was also evident in the contralateral NTS (C) and NA (D). Scale bars= 30 mm.
Fig. 4. A diagramatic representation of the caudal medulla in cross-section identifying areas of interest for this study (Adaptation from Paxinos and Watson (1997)) and illustrating the caudo-rostral pattern of fluorescence at 48 h following tracer administration. Sections from left to right are: (1) the most caudal representation of the NTS in the medulla; (2) the obex of the 4th ventricle; (3) the most rostral extent of the area postrema; and (4) 1.5 mm rostral to the obex of the 4th ventricle. Only areas with a fluorescence intensity ratio greater than 1.0 are indicated. c : areas with autofluorescence; *: areas of fluorescence in images with labelled neurons; NTS: Nucleus tractus solitarius; NA: Nucleus ambiguus; DMV: Dorsal motor nucleus of the vagus; HG: Hypoglossal nucleus; VMRN: Ventral medullary reticular nucleus.
144
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
Table 2 Mean dimensions and numbers of WGA–TRITC labeled neurons (n = 8)
Number of cells with nuclei/3.6 mm2 Length of major cell axis (mm) Length of minor cell axis (mm) Cell surface area (mm2)
NTSa (SD) [Range]
NAb (SD) [Range]
DMVc (SD) [Range]
HGd (SD) [Range]
5 (2) 18.1 (4.4) [13.9–28.2] 9.7 (2.8) [6.3–14.5] 137.2 (60.1) [81.2– 268.9]
4 (2) 26.2 (5.9) [14.2–35.3] 14.4 (3.4) [9.0–19.9] 264.2 (84.5) [91.0–347.9]
7 (4) 18.3 (3.9) [12.3–23.7] 11.0 (2.2) [8.1–14.0] 159.3 (56.2) [105.9–247.5]
3 (1) 30.0 (4.8) [20.8–36.6] 15.3 (2.2) [11.2–18.2] 314.5 (80.5) [165.1–413.8]
a
NTS: Nucleus tractus solitarius, NA: Nucleus ambiguus, c DMV: Dorsal nucleus of the vagus, d HG: Hypoglossal nucleus. b
Intense fluorescence label was visible in the ipsilateral NA from about 250 mm caudal to the obex to 2 mm rostral to the obex, and included populations of neurons in the inferior olive, intermediate reticular nucleus and gigantocellular reticular nucleus. Additionally, a population of fluorescent neurons was evident in the more rostral spinal trigeminal nucleus interpolaris.
3.3. Number and size of labeled neurons The number of cells labeled and their dimensions for each brainstem projection site is shown in Table 2. Neurons of the DMV were most numerous whereas NA and HG neurons were sparse and NTS neurons were intermediate in number. There was no significant difference in axis length between nuclei of smaller neurons (NTS, DMV), nor was there a difference in surface area between the two. Also, there was no significant difference between the groups having larger neurons in these dimensions (NA, HG neurons). Overall, there was a significant difference in all dimensions between groups having smaller and larger neurons (P B 0.01). There was no apparent relationship between cell dimension and fluorescence intensity.
3.4. Distribution of autofluorescence in the caudal brainstem The caudo-rostral distribution of autofluorescence is presented in Fig. 4 (c symbols). In the most caudal images, there was no autofluorescence in the NA but there was clearly autofluorescence in the large population of neurons of the bilateral caudoventrolateral and lateral reticular nuclei in addition to the widely-spaced giant cells in the ventral medullary reticular nucleus (VMRN). Autofluorescence was identified in the NA, VMRN and inferior olive at the level of the obex. FI of neurons was similar in the three areas. These populations were characteristic of the autofluorescence between the obex and 1.5 mm rostral to the obex at which point the autofluorescence of the NA and VMRN disappeared. Only the autofluorescing giant cells of the
gigantocellular reticular nucleus and sparsely distributed single neurons throughout the spinal nucleus of the trigeminal tract remained. Autofluorescence was scant in the NTS (Fig. 2A) where the FI was 0.849 0.25 but was clearly visible in other locations of the caudal brainstem such as the NA (2.359 0.83; Fig. 2B) and surrounding ventral nuclei, albeit at significantly lower levels than the fluorescence resulting from label (P B 0.01, Table 1). There was no significant difference in FI between the two control groups. Our qualitative observation suggested that when fluorescence intensity ratios were B1.00, fluorescence was marginally discernible.
4. Discussion WGA–TRITC is a novel tracer of peripheral nerve brainstem projections that combines the advantages of lectin-conjugates with fluorescent tracers. We used traditional microscopic techniques combined with computer image analysis to identify SLN and PhN brainstem projection neurons labeled with WGA– TRITC. Fluorescently labeled neurons were located in the ipsilateral NTS, NA, ventral NA, DMV, and HG after 48 h survival time and in the contralateral nuclei by 72 h. Cross-sections of intra-nuclear neuron processes and primary afferents were also visible in the NTS. A confounding problem was brainstem autofluorescence which was scant in the NTS, but clearly visible in the caudal ventral medulla. Although labeled neurons were readily discernible, calculation of a fluorescence intensity ratio was the most reliable quantitative measure of the qualitative difference between the autofluorescence of controls and the fluorescence of labeled neurons.
4.1. In6estigations with lectin-conjugates The trans-neuronal movement of the WGA lectin has been characterized by Broadwell and Balin (1985) in olfactory neurons. They identified the neuronal path-
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
way of WGA in its conjugated form with HRP from adsorptive endocytosis through packaging and exocytosis in transfer vesicles. This packaging and transport process suggests that the WGA conjugate could be useful as a vital intracellular stain for electrophysiology studies. Also, the high affinity of WGA for glycoproteins and glycolipids make it more sensitive than the tracer alone for studying both retrograde and anterograde pathways (Bentivoglio and Chen, 1993). To evaluate the efficacy of WGA – TRITC transfer into the brainstem, we compared the labeling with results from other investigators who utilized WGA– HRP to study brainstem projections from the pharynx and larynx (Altschuler et al., 1989; Mrini and Jean, 1995; Furusawa et al., 1996). Since WGA – HRP is both an anterograde and retrograde tracer, afferent projections in the NTS and efferents in the NA are identified. Although WGA–HRP is useful for tracing brainstem neurons that receive projections from peripheral nerves, it is also problematic because there are secondary projections between the NTS and NA (Hayakawa et al., 1998) that could confuse the interpretation of the labeling pattern. Nonetheless, previous studies indicate that the afferent SLN brainstem projection in the rat is confined to the interstitial nucleus of the NTS (Altschuler et al., 1989; Mrini and Jean, 1995; Furusawa et al., 1996) which extends from slightly caudal to the obex rostrally for approximately 600 – 800 mm. Rat SLN efferents are localized to a 1 mm segment of the semicompact formation of the NA located between 1.5 and 2.5 mm rostral to the obex (Furusawa et al., 1996). The PhN motoneuron pool is caudal to this area of the NA (Beiger and Hopkins, 1987). In our studies, WGA – TRITC label was clearly distributed in the primary SLN and PhN brainstem projection sites of the NTS, NA, and DVM by 48 h after tracer application. Ipsilateral label was most intense in the NA and ventrolateral nuclei from the obex to the mid-fourth ventricle area whereas label in the NTS was strongest in the lateral region of the interstitial subnucleus, but also evident in the medial regions and in the intermediate subnucleus. By 72 h of incubation, label was identified in the contralateral NTS and NA. Our results suggest that 48 h may be longer than the transport time required for WGA – TRITC to label only primary projections because hypoglossal motoneurons which are likely secondary projection sites (Furusawa et al., 1996) were also labeled in our study. This possibility is supported by the extensive labeling in the NTS which should have been confined to the lateral aspect of the interstitial nucleus but was found to include areas rostral and caudal to the interstitial nucleus and the medial aspect of these nuclei. Label in these sites could only be from a trans-neuronal source because we did not label the glossopharyngeal nerve which projects to the intermediate subnucleus. Although the WGA–
145
TRITC label could have moved transynaptically from the interstitial subnucleus, a more likely source is the subcompact formation of the NA because it receives monosynaptic (primarily inhibitory) inputs from the intermediate and interstitial subnucleus of the NTS (Hayakawa et al., 1998). Furthermore, results from one animal with a 42 h survival time, had the label confined to the interstitial subnucleus of the NTS, thus providing additional evidence for using a shorter survival time for labeling.
4.2. Fluorescent tracers of peripheral brainstem projections The advantages of using fluorescent tracers are: (1) The label is immediately visible without post-processing the tissue, thus preserving stain potentially lost in processing and decreasing potential damage to the tissue from continued manipulation (Schmued, 1994; Richmond et al., 1994); (2) Multiple tracers can be used to label adjacent anatomic sites in the same brain slice allowing simultaneous visualization of the different projections (Horner and Kummel, 1993; Richmond et al., 1994); and (3) Direct visualization allows fluorescent tracers to be used as vital stains, identifying specific neurons for in vitro electrophysiology experiments (Viana et al., 1990). Disadvantages of fluorescent tracers are their deterioration and tendency to fade over time (although some fluorescent tracers are amenable to immunocytochemical techniques and/or counter-staining), their decreased sensitivity compared with immunoperoxidase, and the potential confusion of fluorescent label with neuronal autofluorescence (Bentivoglio and Chen, 1993; Schmued, 1994). For our study of brainstem projection neurons, we sought to capitalize on the advantages of fluorescent tracers and were able to overcome the potential disadvantages by using digital photography and computerized image analysis. Because there is little reference made in the literature to brainstem autofluorescence, it came as a surprise to us when our early results were confounded with autofluorescence in and around the NA. Autofluorescence is typically increased with animal age and has been proposed to be caused by the accumulation of lipofuscin in aging cells (Mochzuki et al., 1995) although there may also be other sources of autofluorescence (Bentivoglio and Chen, 1993). By creating a fluorescence intensity ratio that normalized our measurements of fluorescence with respect to the background, we were able to quickly and reliably quantitatively compare the fluorescence between images. Autofluorescence was most apparent in the nuclei ventral to the caudal NA whereas fluorescent label was more concentrated in the NA area and extended from a few slices caudal to the obex, rostrally to the midfourth ventricle area.
146
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
Since fluorescent tracers have not previously been used to trace primary projection neurons from peripheral nerves, we compared our measurements of somata dimensions with published data using other tracers. The mean lengths of the major and minor axis for NTS neurons in our study are within the range determined by Kawai and Senba (1996) with biocytin. They described two types of caudal NTS neurons: smaller, local circuit cells with a major axis of 13.7 9 1.9 mm and a minor axis of 10.991.2 mm and larger projection cells with a major axis of 18.6 94.2 mm and a minor axis of 13.0 92.2 mm. Our labeled neurons which were 18.19 4.4 mm in the major axis and 9.792.8 mm in the minor axis are closer in size to the larger projection cells described by Kawai and Senba. The difference in axis dimensions may be due to artifacts related to tissue postprocessing, including dehydration and clearing methods that tend to shrink tissues. Altschuler et al. (1991) used cholera toxin-HRP to characterize pharyngeal motoneurons in the NA semicompact formation into 3 shapes: multipolar (20 – 30×25 – 35 mm), oval (15 – 20×20–25 mm), and fusiform (15× 30 mm). Although we did not separate the shape of our NA neurons, they were in the mid-size range (26.2 9 5.9 mm by 14.49 3.4 mm).
5. Summary WGA–TRITC appears to be reliably transported from peripheral nerves into brainstem neurons, presumably via anterograde movement from the nodose ganglion and across synapses into neurons of the caudal NTS. It also appears to be transported retrogradely into neurons of the caudal NA. Furthermore, there is some evidence that, with increased survival times, WGA–TRITC moves transynaptically into ipsilateral DMV and HG neurons and into the contralateral NTS. The similarity in size, number, and character of our labeled neurons with data from previous investigations supports our contention that WGA – TRITC is a reliable, albeit less robust, alternative for tracing brainstem projection neurons.
Acknowledgements We are indebted to Dr James M. Tepper and his laboratory personnel at Rutgers University, Newark who provided assistance and guidance during the early phases of this work. We also thank Dr Kris Mosier for her contributions to the image analysis and Rich Marino of Empire Imaging for providing the necessary technical support. This research was supported by NIH grant NS01650 and by a grant from the Foundation of UMDNJ. Preliminary accounts of some of these data
have been previously presented (Sawczuk et al., 1997; Sawczuk and Covell, 1998, 1999).
References Alonso-Varona A, Calle Y, Palomares T, Castro B, Barbera-Guillem E. A simple cell labeling technique by means of lectins linked to fluorochromes for the detection of cells on tissue sections. Biol Cell 1995;83(1):87 – 92. Altschuler SM, Bao X, Bieger D, Hopkins DA, Miselis RR. Viscerotopic representation of the upper alimentary tract in the rat: Sensory ganglia and nuclei of the solitary and spinal trigeminal tracts. J Comp Neurol 1989;283:248 – 68. Altschuler SM, Bao X, Miselis RR. Dendritic architecture of nucleus ambiguus motoneurons projecting to the upper alimentary tract in the rat. J Comp Neurol 1991;309:402 – 14. Beiger D, Hopkins DA. Viscerotopic representation of the upper alimentary tract in the medulla oblongata in the rat: the nucleus ambiguus. J Comp Neurol 1987;262:546 – 62. Bentivoglio M, Chen S. Retrograde neuronal tracing combined with immunocytochemistry. In: Cuello AC, editor. Immunohistochemistry II, IBRO Handbook Series: Methods in the Neurosciences, 14. Chichester: Wiley, 1993:312 – 8. Broadwell RD, Balin BJ. Endocytic and exocytic pathways of the neuronal secretory process and trans-synaptic transfer of wheat germ agglutinin-horseradish peroxidase in vivo. J Comp Neurol 1985;242:632 – 50. Dobbins EG, Feldman JL. Differential innervation of protruder and retractor muscles of the tongue in rat. J Comp Neurol 1995;357:376 – 94. Furusawa K, Yasuda K, Okuda D, Tanaka M, Yamaoka M. Central distribution and peripheral functional properties of afferent and efferent components of the superior laryngeal nerve: morphological and electrophysiological studies in the rat. J Comp Neurol 1996;375:147 – 56. Haugland RP. Handbook of Fluorescent Probes and Research Chemicals. 6th ed. Eugene, OR: Molecular Probes, Inc., 1996. Hayakawa T, Zheng JQ, Seki M, Yajima Y. Synaptology of the direct projections from the nucleus of the solitary tract to pharyngeal motoneurons in the nucleus ambiguus of the rat. J Comp Neurol 1998;393:391 – 401. Horner M, Kummel H. Topographical representation of shoulder motor nuclei in the cat spinal cord as revealed by retrograde fluorochrome tracers. J Comp Neurol 1993;335:309 – 19. Kawai Y, Senba E. Organization of excitatory and inhigitory local networks in the caudal nucleus of tractus solitarius of rats revealed in the in vitro slice preparation. J Comp Neurol 1996;373:309 – 21. Medda BK, Lang IM, Layman R, Hogan WJ, Dodds WJ, Shaker R. Characterization and quantification of a pharyngo-UES contractile reflex in cats. Am J Physiol 1994;267:G972– 83. Menghi G, Marchetti L, Bondi AM, Materzaai G. Contribution of confocal laser scanning microscopy to glycochemistry of mouse and rat submandibular glands by single and double lectin staining. Euro J Histochem 1997;41(2):91 – 104. Mrini A, Jean A. Synaptic organization of the interstitial subdivision of the nucleus tractus solitarii and of its laryngeal afferents in the rat. J Comp Neurol 1995;355:221 – 36. Mochzuki Y, Park MK, Mori T, Kawachima S. The difference in autofluorescence features of lipofuscin between brain and adrenal. Zoolog Sci 1995;12(3):283– 8. Naito J, Ishikawa H, Ohmori Y, Watanabe T. Cells of origin of spinal projections from the paraventricular nucleus (PVN) of the hypothalamus in the chicken. J Auton NS 1996;61(3):321–3.
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147 Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates, compact 3rd ed. San Diego: Academic Press, Inc., 1997. Richmond FJR, Gladdy R, Creasy JL, Kitamura S, Smits E, Thomson DB. Efficacy of seven retrograde tracers, compared in multiple-labeling studies of feline mononeurones. J Neurosci Methods 1994;53:35 – 46. Sawczuk A, Theodorou L, Covell DA. Use of WGA-TRITC to trace superior laryngeal nerve brainstem projections. J Dent Res 1997;77:292. Sawczuk A, Covell DA. Identification of rat superior laryngeal nerve
.
147
brainstem projections labeled with wheat germ agglutinin congujated to rhodamine. Soc Neurosci Abstr 1998;24:1150. Sawczuk A, Covell DA. Densitometry analysis of rat brainstem neurons labeled with WGA-TRITC. J Dent Res 1999;78:358. Schmued LC. Anterograde and retrograde neuroanatomical tract tracing with fluorescent compounds. Neurosci Protocols 1994;5:201 – 15. Viana F, Gibbs L, Berger AJ. Double- and triple-labeling of functionally characterized central neurons projecting to peripheral targets studied in vitro. Neuroscience 1990;38:829 – 41.
Wheat germ agglutinin conjugated to TRITC: A novel approach for labeling primary projection neurons of peripheral afferent nerves Andrea Sawczuk a,c,*, David A. Covell, Jr. b a
Departments of Oral Pathology, Biology, and Diagnostic Sciences, Uni6ersity of Medicine and Dentistry of New Jersey, 1110 Bergen Street, Newark, NJ 07103, USA b Department of Orthodontics, Uni6ersity of Medicine and Dentistry of New Jersey, 1110 Bergen Street, Newark, NJ 07103, USA c Department of Neuroscience, Uni6ersity of Medicine and Dentistry of New Jersey, 1110 Bergen Street, Newark, NJ 07103, USA Received 11 March 1999; received in revised form 12 August 1999; accepted 20 August 1999
Abstract Wheat germ agglutinin conjugated to tetramethylrhodamine isothiocyanate-dextran (WGA – TRITC) was studied as a novel tracer of primary projection neurons of pharyngeal (PhN) and superior laryngeal (SLN) branches of the vagus nerve. The SLN and PhN were dissected from rat cervical tissues and the proximal end of the nerves were bathed in tracer for 60 – 90 min. The animals were sacrificed 42–72 h later. The tissue was fixed, sliced, mounted on slides and viewed under epifluorescence. The clarity of the fluorescent label in projection neurons was confounded in some regions of the brainstem by autofluorescence. A computer image analysis method was developed to quantify fluorescence intensity for definitive identification of labeled neurons. Brainstem neurons labeled by afferent projections of the SLN and PhN were localized to the nucleus tractus solitarius. Efferents were identified in the nucleus ambiguus. WGA–TRITC labeled cells were observed in the ipsilateral brainstem at intensities significantly different from the fluorescence observed in controls (PB 0.01). The distribution and density of labeling is in agreement with results of previous investigations, suggesting that WGA – TRITC is a useful alternative for tracing SLN and PhN projections to brainstem nuclei. © 1999 Elsevier Science B.V. All rights reserved. Keywords: WGA – TRITC; Fluorescent tracers; Lectin conjugates; Autofluorescence; Vagus nerve; Superior laryngeal nerve; Pharyngeal nerve; Nucleus tractus solitarius; Nucleus ambiguus
1. Introduction Wheat germ agglutinin conjugated to horse-radish peroxidase (WGA– HRP) is the preferred tracer of peripheral nerve central projections and pathways (Altschuler et al., 1989; Mrini and Jean, 1995; Furusawa et al., 1996) because the WGA lectin facilitates transynaptic movement of the HRP label into brainstem neurons (Broadwell and Balin, 1985; Beiger and Hopkins, 1987). Recently, WGA conjugated to the fluorescent tracer, tetramethylrhodamine isothiocyanate-dextran (WGA – TRITC) has become commercially available. Although WGA – TRITC has been * Corresponding author. Tel.: +1-973-972-2454; fax: + 1-973-9723164. E-mail address: [email protected] (A. Sawczuk)
routinely applied to non-neuronal cell biology investigations (Alonso-Varona et al., 1995; Rev. in Haugland, 1996; Menghi et al., 1997), this tracer has had limited use in neuroscience (Naito et al., 1996) and has not previously been used to trace peripheral nerve brainstem projections. Fluorescent tracers have proven to be powerful tools for identifying neuronal pathways (Bentivoglio and Chen, 1993; Horner and Kummel, 1993; Richmond et al., 1994; Schmued, 1994) and their many applications could be further expanded to include transynaptic movement if used as a conjugate with WGA. We desired the combined lectin-fluorophore properties of WGA–TRITC to identify primary brainstem projection neurons of superior laryngeal (SLN) and pharyngeal branches of the vagus nerve (PhN). Afferents of the SLN and PhN are known to project via the nodose ganglion to neurons of the nucleus tractus
0165-0270/99/$ - see front matter © 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 0 2 7 0 ( 9 9 ) 0 0 1 4 1 - 7
140
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
solitarius (NTS; Mrini and Jean, 1995; Furusawa et al., 1996). Efferents project from motoneurons in the caudal nucleus ambiguus (NA) to the muscles of the pharynx and larynx (Medda et al., 1994). Although these nerves provide peripheral neural control for swallowing and respiration, their central pathways segregate functionally in the NTS. This functional segregation suggests that respiration and swallowing can be studied independently once the projection neurons have been identified and characterized (Dobbins and Feldman, 1995). The primary purpose of this study was to determine if WGA–TRITC could be used to identify SLN and PhN brainstem projection neurons and could provide sufficient morphological detail of the labeled neurons to be useful as a vital stain. To accomplish this goal, we required a fluorescent label that was sufficiently robust to be discriminated from brainstem autofluorescence. A method for image analysis was developed to quantitatively distinguish labeled neurons from background autofluorescence and to measure the size of labeled neurons.
2. Materials and methods
2.1. Tracer application Adult Sprague–Dawley rats (Taconic; n = 18) weighing 200–450 g were anesthetized by an intraperitoneal injection of ketamine and xylazine (80 and 12 mg/kg, respectively). A midline incision was made from the mid-trachea to the mandible under aseptic conditions. All remaining dissection was performed on the right side only. The submandibular glands and muscles overlying the trachea were gently moved aside to expose the trachea at the level of the larynx and hyoid bone. The posterior belly of the digastric muscle was detached from the hyoid bone and the hyoid bone was lifted to expose the carotid sheath. The superior laryngeal and pharyngeal nerve branches of the vagus were identified beneath the hyoid bone and 1 cm segments of nerve were dissected from the underlying tissue and the carotid sheath. The nerves were cut at their most distal unbranched aspects and the proximal ends were crushed and isolated with a paraffin cuff that was secured to the proximal nerves with a droplet of cyanoacrylate glue. The cut nerve ends were bathed in 3 – 5 ml of 5% WGA – TRITC in saline (Sigma). Salinesoaked cotton was placed over the nerves and exposed muscles during this time to prevent dehydration. After 60 – 90 min, the remaining tracer was carefully rinsed from the nerves, the paraffin with the adherent cyanoacrylate was removed, the surrounding tissue was inspected for tracer leakage, and the surgical site was lavaged with saline. Since the tracer solution was pink
in color, it was readily identified in the surrounding tissue. Data were not collected from any animal with suspected tracer leakage. Following saline irrigation, the nerves, muscles, and glands were replaced in their original positions and the incision was closed with wound clips. Control animals (n = 6) were divided equally into 2 groups. One group received sham surgeries in which the procedures described above were performed but the tracer was not applied. Animals in the other control group were perfused but no surgeries were performed.
2.2. Perfusion and histology Because WGA–TRITC had not been previously applied to peripheral nerves, we estimated minimum survival times from previous studies with WGA–HRP to be about 48 h. Thus, our survival times ranged between 42 and72 h. (42 h: n=1; 48 h: n=2; 54 h: n = 1; 62 h: n= 2; 66 h: n= 1; 72 h: n= 1). Animals were sacrificed with an overdose of sodium pentobarbital (\ 100 mg/ kg) followed immediately by cardiac perfusion with 50 ml of sodium phosphate buffer and then by 150–200 ml of 4% paraformaldehyde in 0.15 M phosphate buffer at pH 7.2. The brainstems were removed and stored in 4% paraformaldehyde in 0.15 M phosphate buffer with 20% sucrose for about 12 h when they were transferred to 0.15 M phosphate buffer (pH 7.4) with 30% sucrose for an additional 12 h. The brainstems were frozen in a cryostat microtome and sectioned into 40 mm slices. The slices were washed 3–5 times in 0.15 M phosphate buffer at pH 7.4 and placed on slides to dry overnight in the dark. The slides were then rinsed briefly in xylene (B 2.5 s) and coverslipped with a nonfluorescent medium (DPX mountant, Fluka). All animal use was in accordance with the PHS manual ‘Guide to the Care and Use of Laboratory Animals’ and was approved by the Institutional Animal Care and Use Committee of the University of Medicine and Dentistry of New Jersey.
2.3. Data analysis Sections were viewed under a compound light microscope (Olympus BX50WI) using a 40× objective with epifluorescence (excitation wavelength of 525–560 nm). The images were captured with a video camera (DageMTI CCD-300-RC) and digitized with a frame grabber board (LG-3 NuBus, Scion Corporation, Frederick, MD) at consistent microscope, camera, and computer settings. The final images were an integrated average of 16 captured images. Fluorescence intensity was calculated using Scion Image computer software (modified from NIH Image) by applying the following protocol to the unfiltered image: (1) To measure fluorescence intensity of cells
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
within a captured image, the background was subtracted from the signal by manually thresholding the inverted image so that only the punctate areas representing fluorescence were counted (i.e. as pixel gray values); (2) Because SLN projections to the NTS were primarily observed as increased fluorescence of processes interspersed with cells in the interstitial nucleus,
141
measurements of the processes were made within a circle 15 mm in diameter using the manual thresholding technique described above; (3) To test for reliability, 10% of the manually thresholded data were randomly chosen, remeasured by a second blinded investigator and compared with the originally measured data using a paired t-test. There was no significant difference between investigators; (4) To normalize the data for between image comparisons, a fluorescence intensity ratio (FI) was calculated relating signal density to background density using the following formula: Cell Signal Density FI= (Cell Background Density + General Background Density)
Cell signal and background values were determined by averaging three measurements of the same image. General background was calculated as the mean thresholded value of 3 squares, 15 mm/side, taken from representative background areas of the image but avoiding cells, cell processes, and black holes (Fig. 1). Uptake of tracer was considered successful only if FI was determined with an unpaired t-test to be significantly greater than the autofluorescence of controls. Characteristics selected to estimate label acceptability were cell numbers, somata dimensions, and label distribution. The criteria for acceptability were based on the results of previous studies using WGA–HRP and HRP because fluorescent tracers have not previously been used to study the brainstem projection of cervical vagal nerves. Cell number was counted within a 3.6 mm2 area in those locations demonstrating tracer. Cell size was calculated by averaging the lengths of the major and minor axis of 3 cell somata. Surface area was determined with automated ROI (Region of Interest) analysis features (Scion Image software). Only cells with nuclei were included in the cell count and morphometry. Label distribution was estimated by examining the label pattern in individual neurons and by identifying brainstem projection sites. Fig. 1. Isolation method of neurons and background areas for calculation of fluorescence intensity ratio. A: Image of labeled NTS neurons and processes after a 48 h incubation period (light gray and white areas indicate fluorescent label or autofluorescence). Scale bar= 30 mm. B: The areas to be analyzed in Panel A have been outlined in white: neurons, areas of cell processes (circles), and background (squares). C: The image has been inverted which causes the fluorescent label to appear dark gray or black and the background to appear light gray or white. A neuron of interest has been outlined in white. D: The density slice option in the image analysis program (Scion Image) has been applied and the exposure level has been manually adjusted to a level that preserves the dark gray and black punctate areas (presumed fluorescence label) in the neuron being studied. Other areas in the field having the same intensity are also visible. E: The image threshold has been generated based on the exposure level established in Panel D. The mean integrated density analysis was calculated from the thresholded image for signal and background. Calculation of the fluorescence intensity ratio is discussed in Section 2.3. Scale bar = 15 mm.
3. Results
3.1. Identification of labeled projection neurons Label was clearly located in the NTS, NA, ventral NA, DVM, and HG neurons by 48 h survival time. Fluorescence was discernible in the cell somata and proximal processes (Fig. 2C, D) as pink to red granular inclusions primarily concentrated in the cytoplasm surrounding unstained nuclei and more sparsely distributed in cell processes. Label was also visible as small clumps without distinct nuclei (Fig. 2C) in the solitary tract. These clumps were suggestive of cell processes in cross-section and probably included intra-nuclear neuronal processes and primary afferent nerves.
142
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
Fig. 2. Relative difference in fluorescence intensity can be seen in photomicrographs of NTS (A) and NA (B) from control animals compared to the NTS (C) and NA (D) from labeled animals. The fluorescence intensity ratio for labeled neurons of the NTS was approximately 4 times the fluorescence intensity of control neurons (A vs C) and for labeled neurons of the NA was 3.5 times the ratio of control neurons (B vs D). Scale bar= 30 mm. Table 1 Measurements of Ipsilateral Fluorescence Intensity Ratio (FI)a
Control cell autofluorescence WGA–TRITC labeled cells Control processes autofluorescence WGA–TRITC labeled processes
NTSb (SD)
NAc (SD)
DMVd (SD)
HGe (SD)
0.84 2.53 0.65 2.93
2.35 (0.91) 6.63 (2.89)** – –
1.58 (1.13) 2.34 (1.40)* – –
4.18 (1.35) 9.38 (4.29)** – –
(0.50) (0.90)*** (0.54) (2.23)*
a
Values in parenthesis: Ratio of mean pixel gray values of somata or processes to background. NTS: Nucleus tractus solitarius, c NA: Nucleus ambiguus, d DMV: Dorsal nucleus of the vagus, e HG: Hypoglossal nucleus. * PB0.05, ** PB0.01, *** PB0.001. b
Autofluorescence was also granular and had a similar cytoplasmic distribution to labeled neurons but was yellowish in color (Fig. 2A, B). Although the color was generally different between autofluorescence and label, the most dependable distinguishing feature was fluorescence intensity. The intensity of the WGA– TRITC label was quantitatively verified by the computer image analysis that showed a significant elevation in fluorescence intensity relative to the autofluorescence of controls (P B 0.01). FI of control NTS cells in Fig. 2 was 0.72 and for processes was
1.31, whereas the FI of label in NTS somata (3.33) and processes (7.08) was significantly greater than controls (PB 0.001, Table 1). The FI of control and labeled NA neurons in Fig. 2 was also significantly different (PB 0.01, Table 1). Additionally, tracer was identified in the ipsilateral DMV and HG at intensity levels significantly elevated from controls (Table 1). By 72 h tracer appeared to have moved beyond the primary projection neurons (Fig. 3), labeling the contralateral NTS (FI= 1.68) and NA (FI= 11.00). Tracer was also seen in the contralateral DMV but not in the contralateral HG by 72 h.
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
3.2. Brainstem distribution of labeled projection neurons Fig. 4 shows the distribution of label in cross-sectional drawings from the most caudal to the most rostral aspect of brainstem projections. Fluorescence label was visible in the ipsilateral NTS by 48 h of incubation (* symbols in Fig. 4) from about 3 slices
143
caudal to the obex (120 mm) to 2 mm rostral to the obex. Although fluorescence was most intense in somata and processes of the interstitial subnucleus, rostral areas of the NTS compatible with the intermediate nucleus were also labeled. Fluorescence was most concentrated in the lateral NTS, primarily in processes, and less concentrated in the medial aspect of the nucleus.
Fig. 3. Fluorescence label at 72 h in the ipsilateral NTS (A) and NA (B). Label was also evident in the contralateral NTS (C) and NA (D). Scale bars= 30 mm.
Fig. 4. A diagramatic representation of the caudal medulla in cross-section identifying areas of interest for this study (Adaptation from Paxinos and Watson (1997)) and illustrating the caudo-rostral pattern of fluorescence at 48 h following tracer administration. Sections from left to right are: (1) the most caudal representation of the NTS in the medulla; (2) the obex of the 4th ventricle; (3) the most rostral extent of the area postrema; and (4) 1.5 mm rostral to the obex of the 4th ventricle. Only areas with a fluorescence intensity ratio greater than 1.0 are indicated. c : areas with autofluorescence; *: areas of fluorescence in images with labelled neurons; NTS: Nucleus tractus solitarius; NA: Nucleus ambiguus; DMV: Dorsal motor nucleus of the vagus; HG: Hypoglossal nucleus; VMRN: Ventral medullary reticular nucleus.
144
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
Table 2 Mean dimensions and numbers of WGA–TRITC labeled neurons (n = 8)
Number of cells with nuclei/3.6 mm2 Length of major cell axis (mm) Length of minor cell axis (mm) Cell surface area (mm2)
NTSa (SD) [Range]
NAb (SD) [Range]
DMVc (SD) [Range]
HGd (SD) [Range]
5 (2) 18.1 (4.4) [13.9–28.2] 9.7 (2.8) [6.3–14.5] 137.2 (60.1) [81.2– 268.9]
4 (2) 26.2 (5.9) [14.2–35.3] 14.4 (3.4) [9.0–19.9] 264.2 (84.5) [91.0–347.9]
7 (4) 18.3 (3.9) [12.3–23.7] 11.0 (2.2) [8.1–14.0] 159.3 (56.2) [105.9–247.5]
3 (1) 30.0 (4.8) [20.8–36.6] 15.3 (2.2) [11.2–18.2] 314.5 (80.5) [165.1–413.8]
a
NTS: Nucleus tractus solitarius, NA: Nucleus ambiguus, c DMV: Dorsal nucleus of the vagus, d HG: Hypoglossal nucleus. b
Intense fluorescence label was visible in the ipsilateral NA from about 250 mm caudal to the obex to 2 mm rostral to the obex, and included populations of neurons in the inferior olive, intermediate reticular nucleus and gigantocellular reticular nucleus. Additionally, a population of fluorescent neurons was evident in the more rostral spinal trigeminal nucleus interpolaris.
3.3. Number and size of labeled neurons The number of cells labeled and their dimensions for each brainstem projection site is shown in Table 2. Neurons of the DMV were most numerous whereas NA and HG neurons were sparse and NTS neurons were intermediate in number. There was no significant difference in axis length between nuclei of smaller neurons (NTS, DMV), nor was there a difference in surface area between the two. Also, there was no significant difference between the groups having larger neurons in these dimensions (NA, HG neurons). Overall, there was a significant difference in all dimensions between groups having smaller and larger neurons (P B 0.01). There was no apparent relationship between cell dimension and fluorescence intensity.
3.4. Distribution of autofluorescence in the caudal brainstem The caudo-rostral distribution of autofluorescence is presented in Fig. 4 (c symbols). In the most caudal images, there was no autofluorescence in the NA but there was clearly autofluorescence in the large population of neurons of the bilateral caudoventrolateral and lateral reticular nuclei in addition to the widely-spaced giant cells in the ventral medullary reticular nucleus (VMRN). Autofluorescence was identified in the NA, VMRN and inferior olive at the level of the obex. FI of neurons was similar in the three areas. These populations were characteristic of the autofluorescence between the obex and 1.5 mm rostral to the obex at which point the autofluorescence of the NA and VMRN disappeared. Only the autofluorescing giant cells of the
gigantocellular reticular nucleus and sparsely distributed single neurons throughout the spinal nucleus of the trigeminal tract remained. Autofluorescence was scant in the NTS (Fig. 2A) where the FI was 0.849 0.25 but was clearly visible in other locations of the caudal brainstem such as the NA (2.359 0.83; Fig. 2B) and surrounding ventral nuclei, albeit at significantly lower levels than the fluorescence resulting from label (P B 0.01, Table 1). There was no significant difference in FI between the two control groups. Our qualitative observation suggested that when fluorescence intensity ratios were B1.00, fluorescence was marginally discernible.
4. Discussion WGA–TRITC is a novel tracer of peripheral nerve brainstem projections that combines the advantages of lectin-conjugates with fluorescent tracers. We used traditional microscopic techniques combined with computer image analysis to identify SLN and PhN brainstem projection neurons labeled with WGA– TRITC. Fluorescently labeled neurons were located in the ipsilateral NTS, NA, ventral NA, DMV, and HG after 48 h survival time and in the contralateral nuclei by 72 h. Cross-sections of intra-nuclear neuron processes and primary afferents were also visible in the NTS. A confounding problem was brainstem autofluorescence which was scant in the NTS, but clearly visible in the caudal ventral medulla. Although labeled neurons were readily discernible, calculation of a fluorescence intensity ratio was the most reliable quantitative measure of the qualitative difference between the autofluorescence of controls and the fluorescence of labeled neurons.
4.1. In6estigations with lectin-conjugates The trans-neuronal movement of the WGA lectin has been characterized by Broadwell and Balin (1985) in olfactory neurons. They identified the neuronal path-
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
way of WGA in its conjugated form with HRP from adsorptive endocytosis through packaging and exocytosis in transfer vesicles. This packaging and transport process suggests that the WGA conjugate could be useful as a vital intracellular stain for electrophysiology studies. Also, the high affinity of WGA for glycoproteins and glycolipids make it more sensitive than the tracer alone for studying both retrograde and anterograde pathways (Bentivoglio and Chen, 1993). To evaluate the efficacy of WGA – TRITC transfer into the brainstem, we compared the labeling with results from other investigators who utilized WGA– HRP to study brainstem projections from the pharynx and larynx (Altschuler et al., 1989; Mrini and Jean, 1995; Furusawa et al., 1996). Since WGA – HRP is both an anterograde and retrograde tracer, afferent projections in the NTS and efferents in the NA are identified. Although WGA–HRP is useful for tracing brainstem neurons that receive projections from peripheral nerves, it is also problematic because there are secondary projections between the NTS and NA (Hayakawa et al., 1998) that could confuse the interpretation of the labeling pattern. Nonetheless, previous studies indicate that the afferent SLN brainstem projection in the rat is confined to the interstitial nucleus of the NTS (Altschuler et al., 1989; Mrini and Jean, 1995; Furusawa et al., 1996) which extends from slightly caudal to the obex rostrally for approximately 600 – 800 mm. Rat SLN efferents are localized to a 1 mm segment of the semicompact formation of the NA located between 1.5 and 2.5 mm rostral to the obex (Furusawa et al., 1996). The PhN motoneuron pool is caudal to this area of the NA (Beiger and Hopkins, 1987). In our studies, WGA – TRITC label was clearly distributed in the primary SLN and PhN brainstem projection sites of the NTS, NA, and DVM by 48 h after tracer application. Ipsilateral label was most intense in the NA and ventrolateral nuclei from the obex to the mid-fourth ventricle area whereas label in the NTS was strongest in the lateral region of the interstitial subnucleus, but also evident in the medial regions and in the intermediate subnucleus. By 72 h of incubation, label was identified in the contralateral NTS and NA. Our results suggest that 48 h may be longer than the transport time required for WGA – TRITC to label only primary projections because hypoglossal motoneurons which are likely secondary projection sites (Furusawa et al., 1996) were also labeled in our study. This possibility is supported by the extensive labeling in the NTS which should have been confined to the lateral aspect of the interstitial nucleus but was found to include areas rostral and caudal to the interstitial nucleus and the medial aspect of these nuclei. Label in these sites could only be from a trans-neuronal source because we did not label the glossopharyngeal nerve which projects to the intermediate subnucleus. Although the WGA–
145
TRITC label could have moved transynaptically from the interstitial subnucleus, a more likely source is the subcompact formation of the NA because it receives monosynaptic (primarily inhibitory) inputs from the intermediate and interstitial subnucleus of the NTS (Hayakawa et al., 1998). Furthermore, results from one animal with a 42 h survival time, had the label confined to the interstitial subnucleus of the NTS, thus providing additional evidence for using a shorter survival time for labeling.
4.2. Fluorescent tracers of peripheral brainstem projections The advantages of using fluorescent tracers are: (1) The label is immediately visible without post-processing the tissue, thus preserving stain potentially lost in processing and decreasing potential damage to the tissue from continued manipulation (Schmued, 1994; Richmond et al., 1994); (2) Multiple tracers can be used to label adjacent anatomic sites in the same brain slice allowing simultaneous visualization of the different projections (Horner and Kummel, 1993; Richmond et al., 1994); and (3) Direct visualization allows fluorescent tracers to be used as vital stains, identifying specific neurons for in vitro electrophysiology experiments (Viana et al., 1990). Disadvantages of fluorescent tracers are their deterioration and tendency to fade over time (although some fluorescent tracers are amenable to immunocytochemical techniques and/or counter-staining), their decreased sensitivity compared with immunoperoxidase, and the potential confusion of fluorescent label with neuronal autofluorescence (Bentivoglio and Chen, 1993; Schmued, 1994). For our study of brainstem projection neurons, we sought to capitalize on the advantages of fluorescent tracers and were able to overcome the potential disadvantages by using digital photography and computerized image analysis. Because there is little reference made in the literature to brainstem autofluorescence, it came as a surprise to us when our early results were confounded with autofluorescence in and around the NA. Autofluorescence is typically increased with animal age and has been proposed to be caused by the accumulation of lipofuscin in aging cells (Mochzuki et al., 1995) although there may also be other sources of autofluorescence (Bentivoglio and Chen, 1993). By creating a fluorescence intensity ratio that normalized our measurements of fluorescence with respect to the background, we were able to quickly and reliably quantitatively compare the fluorescence between images. Autofluorescence was most apparent in the nuclei ventral to the caudal NA whereas fluorescent label was more concentrated in the NA area and extended from a few slices caudal to the obex, rostrally to the midfourth ventricle area.
146
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147
Since fluorescent tracers have not previously been used to trace primary projection neurons from peripheral nerves, we compared our measurements of somata dimensions with published data using other tracers. The mean lengths of the major and minor axis for NTS neurons in our study are within the range determined by Kawai and Senba (1996) with biocytin. They described two types of caudal NTS neurons: smaller, local circuit cells with a major axis of 13.7 9 1.9 mm and a minor axis of 10.991.2 mm and larger projection cells with a major axis of 18.6 94.2 mm and a minor axis of 13.0 92.2 mm. Our labeled neurons which were 18.19 4.4 mm in the major axis and 9.792.8 mm in the minor axis are closer in size to the larger projection cells described by Kawai and Senba. The difference in axis dimensions may be due to artifacts related to tissue postprocessing, including dehydration and clearing methods that tend to shrink tissues. Altschuler et al. (1991) used cholera toxin-HRP to characterize pharyngeal motoneurons in the NA semicompact formation into 3 shapes: multipolar (20 – 30×25 – 35 mm), oval (15 – 20×20–25 mm), and fusiform (15× 30 mm). Although we did not separate the shape of our NA neurons, they were in the mid-size range (26.2 9 5.9 mm by 14.49 3.4 mm).
5. Summary WGA–TRITC appears to be reliably transported from peripheral nerves into brainstem neurons, presumably via anterograde movement from the nodose ganglion and across synapses into neurons of the caudal NTS. It also appears to be transported retrogradely into neurons of the caudal NA. Furthermore, there is some evidence that, with increased survival times, WGA–TRITC moves transynaptically into ipsilateral DMV and HG neurons and into the contralateral NTS. The similarity in size, number, and character of our labeled neurons with data from previous investigations supports our contention that WGA – TRITC is a reliable, albeit less robust, alternative for tracing brainstem projection neurons.
Acknowledgements We are indebted to Dr James M. Tepper and his laboratory personnel at Rutgers University, Newark who provided assistance and guidance during the early phases of this work. We also thank Dr Kris Mosier for her contributions to the image analysis and Rich Marino of Empire Imaging for providing the necessary technical support. This research was supported by NIH grant NS01650 and by a grant from the Foundation of UMDNJ. Preliminary accounts of some of these data
have been previously presented (Sawczuk et al., 1997; Sawczuk and Covell, 1998, 1999).
References Alonso-Varona A, Calle Y, Palomares T, Castro B, Barbera-Guillem E. A simple cell labeling technique by means of lectins linked to fluorochromes for the detection of cells on tissue sections. Biol Cell 1995;83(1):87 – 92. Altschuler SM, Bao X, Bieger D, Hopkins DA, Miselis RR. Viscerotopic representation of the upper alimentary tract in the rat: Sensory ganglia and nuclei of the solitary and spinal trigeminal tracts. J Comp Neurol 1989;283:248 – 68. Altschuler SM, Bao X, Miselis RR. Dendritic architecture of nucleus ambiguus motoneurons projecting to the upper alimentary tract in the rat. J Comp Neurol 1991;309:402 – 14. Beiger D, Hopkins DA. Viscerotopic representation of the upper alimentary tract in the medulla oblongata in the rat: the nucleus ambiguus. J Comp Neurol 1987;262:546 – 62. Bentivoglio M, Chen S. Retrograde neuronal tracing combined with immunocytochemistry. In: Cuello AC, editor. Immunohistochemistry II, IBRO Handbook Series: Methods in the Neurosciences, 14. Chichester: Wiley, 1993:312 – 8. Broadwell RD, Balin BJ. Endocytic and exocytic pathways of the neuronal secretory process and trans-synaptic transfer of wheat germ agglutinin-horseradish peroxidase in vivo. J Comp Neurol 1985;242:632 – 50. Dobbins EG, Feldman JL. Differential innervation of protruder and retractor muscles of the tongue in rat. J Comp Neurol 1995;357:376 – 94. Furusawa K, Yasuda K, Okuda D, Tanaka M, Yamaoka M. Central distribution and peripheral functional properties of afferent and efferent components of the superior laryngeal nerve: morphological and electrophysiological studies in the rat. J Comp Neurol 1996;375:147 – 56. Haugland RP. Handbook of Fluorescent Probes and Research Chemicals. 6th ed. Eugene, OR: Molecular Probes, Inc., 1996. Hayakawa T, Zheng JQ, Seki M, Yajima Y. Synaptology of the direct projections from the nucleus of the solitary tract to pharyngeal motoneurons in the nucleus ambiguus of the rat. J Comp Neurol 1998;393:391 – 401. Horner M, Kummel H. Topographical representation of shoulder motor nuclei in the cat spinal cord as revealed by retrograde fluorochrome tracers. J Comp Neurol 1993;335:309 – 19. Kawai Y, Senba E. Organization of excitatory and inhigitory local networks in the caudal nucleus of tractus solitarius of rats revealed in the in vitro slice preparation. J Comp Neurol 1996;373:309 – 21. Medda BK, Lang IM, Layman R, Hogan WJ, Dodds WJ, Shaker R. Characterization and quantification of a pharyngo-UES contractile reflex in cats. Am J Physiol 1994;267:G972– 83. Menghi G, Marchetti L, Bondi AM, Materzaai G. Contribution of confocal laser scanning microscopy to glycochemistry of mouse and rat submandibular glands by single and double lectin staining. Euro J Histochem 1997;41(2):91 – 104. Mrini A, Jean A. Synaptic organization of the interstitial subdivision of the nucleus tractus solitarii and of its laryngeal afferents in the rat. J Comp Neurol 1995;355:221 – 36. Mochzuki Y, Park MK, Mori T, Kawachima S. The difference in autofluorescence features of lipofuscin between brain and adrenal. Zoolog Sci 1995;12(3):283– 8. Naito J, Ishikawa H, Ohmori Y, Watanabe T. Cells of origin of spinal projections from the paraventricular nucleus (PVN) of the hypothalamus in the chicken. J Auton NS 1996;61(3):321–3.
A. Sawczuk, D.A. Co6ell, Jr. / Journal of Neuroscience Methods 93 (1999) 139–147 Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates, compact 3rd ed. San Diego: Academic Press, Inc., 1997. Richmond FJR, Gladdy R, Creasy JL, Kitamura S, Smits E, Thomson DB. Efficacy of seven retrograde tracers, compared in multiple-labeling studies of feline mononeurones. J Neurosci Methods 1994;53:35 – 46. Sawczuk A, Theodorou L, Covell DA. Use of WGA-TRITC to trace superior laryngeal nerve brainstem projections. J Dent Res 1997;77:292. Sawczuk A, Covell DA. Identification of rat superior laryngeal nerve
.
147
brainstem projections labeled with wheat germ agglutinin congujated to rhodamine. Soc Neurosci Abstr 1998;24:1150. Sawczuk A, Covell DA. Densitometry analysis of rat brainstem neurons labeled with WGA-TRITC. J Dent Res 1999;78:358. Schmued LC. Anterograde and retrograde neuroanatomical tract tracing with fluorescent compounds. Neurosci Protocols 1994;5:201 – 15. Viana F, Gibbs L, Berger AJ. Double- and triple-labeling of functionally characterized central neurons projecting to peripheral targets studied in vitro. Neuroscience 1990;38:829 – 41.