This manuscript describes a protocol to track the re-distribution of branchial ionocytes and their innervation using a time differential staining technique coupled with full bilateral gill denervation.
Branchial ionocytes (ICs) are the functional units for ionic regulation in fish. In adults, they are found on the filamental and lamellar epithelia of the gill where they transport ions such as Na+, Cl– and Ca2+ via a variety of ion channels, pumps and exchangers. The teleost gill is extrinsically innervated by the facial (VI), glossopharyngeal (IX) and vagus (X) nerves. The IX and X nerves are also the extrinsic source of branchial IC innervation. Here, two techniques used to study the innervation, proliferation and distribution of ICs are described: a time differential staining technique and a full bilateral gill denervation technique. Briefly, goldfish are exposed to a vital mitochondrion-specific dye (e.g., MitoTracker Red) which labels (red fluorescence) pre-existing ICs. Fish were either allowed to recover for 3 – 5 days or immediately underwent a full bilateral gill denervation. After 3 – 5 days of recovery, the gills are harvested and fixed for immunohistochemistry. The tissue is then stained with an α-5 primary antibody (targets Na+/K+ ATPase containing cells) in conjunction with a secondary antibody that labels all (both new and pre-existing) ICs green. Using confocal imaging, it was demonstrated that pre-existing ICs appear yellow (labelled with both a viable mitochondrion-specific dye and α-5) and new ICs appear green (labelled with α-5 only). Both techniques used in tandem can be applied to study the innervation, proliferation and distribution of ICs on the gill filament when fish are exposed to environmental challenges.
ICs are the functional unit for ionic regulation in fish and are found on the epithelial surfaces of the gill filaments and lamellae4,6-8,10. Although a variety of subtypes have been described that possess unique features, many of the ICs are characterized by a high density of mitochondria (thus they are also known as mitochondrion-rich cells) and/or an abundance of the enzyme Na+/K+ ATPase (NKA). Typically, these ICs house a variety of other pumps, ion channels and exchangers involved in ion regulation (e.g., Na+/H+ exchanger, Na+/Cl– co-transporter, H+ pump)2,10,11. The redistribution and proliferation of ICs as a compensatory mechanism is central for maintaining ion homeostasis particularly during ionic stress (e.g., exposure to ion-poor water)4,8,9.
This study describes a time differential staining technique1 to identify newly proliferated ionocytes (ICs) in fish gills. This technique is coupled with a complete bilateral denervation of the gill arches. Goldfish (Carassius auratus), the species used in these studies, is well suited to study proliferation of gill epithelial cells because they have a remarkable ability to structurally remodel their gills2. Gill remodeling refers to the growth or retraction of an interlamellar cell mass (ILCM) when the fish (normally maintained at 15 – 30 °C) are acclimated to cold water (<15 °C) or hypoxia, respectively3. Previous studies using the time differential staining technique on goldfish have focused on the redistribution, innervation and proliferation of ICs on the gill in the context of gill remodelling4,5. Katoh and Kaneko1 developed this novel technique to study the transformation and replacement of branchial ICs in killifish (Fundulus heteroclitus) transferred from sea water (SW) to fresh water (FW). In this study, the focus is on the proliferation and innervation of ICs in goldfish acclimated to 25 °C.
Using the time differential staining technique it was shown that, in the context of gill remodelling, goldfish maintain a constant number of ICs during hypoxic exposure and subsequent normoxic recovery, however, the percentage of innervated cells decreased throughout the normoxic recovery period5. It was proposed more than 70 years ago that ionic uptake mechanisms in fish are under neural control12. The teleost gill is innervated by the facial (VII), glossopharyngeal (IX) and vagus (X) nerves also referred to as the “branchial nerves”13,14. Studies by Jonz and Nurse (2003) on zebrafish (Danio rerio) gill innervation showed that the origin of innervation is extrinsic (cell body of nerve fiber is extrinsic to the gill) as well as intrinsic (cell body of nerve fiber is intrinsic to the gill). The same authors also demonstrated that branchial ICs are extrinsically innervated7.
In this study, the time differential staining technique coupled with full bilateral gill denervation was used to investigate the proliferation of ICs lacking extrinsic innervation in goldfish. Full bilateral gill denervation refers to severing cranial nerves IX and X. These two approaches are feasible in goldfish because their relatively large size (from 30 – 200 g) simplifies the delicate surgical procedures, and ionocytes are readily identified using standard immuno-histochemical techniques. In the present study, ICs were visualized using a vital mitochondrion-specific dye (e.g., MitoTracker Red) or a primary antibody against the α-subunit of the Na+/K+-ATPase (α-5; Developmental Studies Hybridoma Bank, University of Iowa, Iowa City IA). This protocol provides a straightforward method of visualizing and analysing the redistribution and proliferation of ICs on the fish gill.
Both protocols conformed to the guidelines of the Canadian Council of Animal Care (CCAC) and were carried out with the approval of the University of Ottawa Animal Care Committee (Protocol BL-226).
1. Time Differential Staining Technique: Mitochondrion-rich Dye Bath
2. Full Bilateral Denervation Procedure
3. Time Differential Staining Technique: Immunohistochemistry
4. Imaging
5. Image Analysis for Ionocyte Quantification
Figure 1 illustrates the surgery table set up (Figure 1A), the placement of the fish during surgery (Figure 1B) and the three most important steps for the time differential staining technique (Figure 1C). In Step 1, the fish is kept for 30 min in a well aerated water bath at 25 °C in the dark. During the 30 min period, the researcher can prepare the mitochondrion-rich dye aliquot in DMSO which is added to the water during Step 2 (Figure 1C). The incubation period in Step 2 allows for uptake of the mitochondrion-rich dye from the water into the mitochondrion-rich cells (i.e., ICs). The fish can then either undergo the full bilateral denervation procedure or a sham procedure in which the fish is anaesthetized and the opercula manipulated but the nerves remain intact. The set up in Step 3 represents a recovery chamber provided with flowing water for a fish that has either undergone full bilateral denervation or a “sham” procedure.
After the recovery period, the fish was euthanized and the gills were excised and fixed for immunohistochemistry. The overall distribution and innervation of ICs on the gill filament of a fish that had undergone a “sham” procedure is depicted in Figure 2. The ICs are present on the filamental epithelia as well as at the base of the interlamellar regions. Figure 2A shows pre-existing ICs labelled with mitochondrion-rich dye (i.e., these ICs existed before the denervation/sham procedures were performed). Figure 2B shows nerve fibres innervating the pre-existing and newly formed ICs (identified by NKA immunoreactivity) of the filamental and lamellar epithelia. Finally, merging of the two images (Figure 2C) clearly reveals the pre-existing ICs (appear yellow) and the new ICs (appear green). Figure 2D is a representative graph of IC quantification for the filament depicted in Figure 2A-C. On this specific filament, there appear to be a larger number of newly proliferated ICs per mm2 than pre-existing ICs (N = 1). To remove the source of extrinsic branchial innervation, the IX and X cranial nerves (extrinsic innervation) were severed. Figure 3A shows the dorsal region of the opercular cavity after the gill filaments and the epithelia covering the nerves were removed to expose the IX and X cranial nerves (indicated with Roman numerals) which span from two major nerve trunks to innervate all four arches (Figure 3A); the gill arches are numbered 1 through 4. Figure 3B-E illustrate selective denervation of the first gill arch. The first gill arch is innervated by both the IX and X cranial nerves (Figure 3B) which can be removed without affecting the innervation to the rest of the gill arches (Figure 3C-E). Full bilateral denervation results in gradual loss of extrinsic innervation to the gill filaments (Figures 4A-C). Control fish exhibit an obvious nerve bundle which spans the length of the filament (Figure 4A). Full bilateral denervation resulted in some loss of extrinsic innervation after 2 days of recovery (Figure 4B). Further disappearance of extrinsic innervation was noted after 5 days of recovery from denervation (Figure 4C). Any remaining innervation to ICs after 5 days of recovery presumably was derived from nerves with cell bodies within the gill filament (intrinsic innervation; Figure 4C).
Figure 1. Experimental set up for the denervation procedure and sequence of steps used in the time differential staining technique. (A) Surgery table set up for denervation procedure showing the anaesthetic and recovery tanks. (B) Example of the placement of an anaesthetized, intubated fish. (C) Key steps used in the time differential staining technique. In Step 1, the fish is placed in a temperature controlled, static aerated water bath for 30 min. The mitochondrion-rich dye is added in Step 2 to a final concentration of 0.1 µM and the fish is allowed to bathe in the solution for a minimum of 4 hr. Water flow is restarted in Step 3 at which point the fish undergoes either a full bilateral denervation of the gill or a sham surgery. Please click here to view a larger version of this figure.
Figure 2. Light micrographs representing the time differential staining technique in a goldfish (acclimated to 25 °C) 2 days after full bilateral denervation. (A) The distribution of pre-existing ionocytes (ICs; arrows) on a single gill filament is revealed by mitochondrion-rich dye staining. (B) The distribution of ICs (new and pre-existing) and branchial nerves (dashed arrows) is revealed by staining with the α-5 and zn-12 antibodies, respectively. The arrows indicate ICs. (C) The overlap of (A) and (B) distinguishes pre-existing ICs (appear yellow; indicated by arrowheads) from newly proliferated ICs (appear green; indicated by arrows). Insert in (C) is a magnification of an innervated pre-existing ionocyte. (D) Representative graph of IC quantification for the filament shown in panels (A-C). N = 1. Scale bar in panel (C) is 50 µm and applies to all panels. Please click here to view a larger version of this figure.
Figure 3. Representative images depicting various stages of the gill innervation. (A) Dorsal view of the buccal cavity showing the IX and X cranial nerves innervating all 4 gill arches. Gill arches are numbered 1-4. The gill filaments and the tissue covering the nerves were removed to better visualize the innervation. (B) The 1st and 2nd gill arches are separated to reveal the sensory organ and the branches of the IX and X cranial nerves innervating the 1st gill arch. (C-E) Sequence of images showing the selective denervation of the 1st gill arch by severing the branches of the IX and X cranial nerves. For a schematic representation of gill innervation of teleost fish, refer to Figure 1 in Milsom et al.26 The white lines in the images are water reflection from the microscope lights; they do not define any morphological structure. Scale bar in panel (E) is 4 mm and applies to all panels. Please click here to view a larger version of this figure.
Figure 4. Light micrographs depicting the distribution and innervation of ionocytes on a single filament of the 1st gill arch of a goldfish acclimated to 25 °C. Ionocytes (indicated by arrow heads) were stained with the α-5 antibody and nerves (indicated by arrows) were stained with the zn-12 antibody. (A) Gill filament of a control fish showing a central nerve bundle presumably originating from the IX and X nerves (extrinsic innervation) with extensive lamellar branching. Some of the ionocytes indicated are also innervated (insert). (B) A gill filament 2 days after full bilateral denervation. There was a reduction of the central nerve bundle while the lamellar innervation appeared largely intact. (C) A gill filament 5 days after full bilateral denervation demonstrating that extrinsic innervation was largely absent. Qualitative analysis suggests that full bilateral denervation causes degradation of the extrinsic gill innervation while maintaining nerves with cell bodies within the gill filament (intrinsic innervation) creating a network of nerves through the filament and into the lamellae. Please click here to view a larger version of this figure.
The time differential staining technique can be a useful tool to understand the dynamic regulation of ion uptake and to examine the temporal redistribution of ICs in the gill epithelia. Although a straightforward procedure, there are a number of key points that are crucial to the success of the time differential staining technique. The goldfish must be exposed to the mitochondrion-rich dye for the time allotted in the protocol. Shorter exposures will result in poor uptake of the dye by the mitochondrion-rich cells (i.e., ionocytes). During fixation, the gill tissue must be excised quickly and kept in the dark to avoid photo bleaching. The tissue must be processed for imaging within 2 weeks of fixation. During the bilateral nerve sectioning procedure ensure that the fish is well anaesthetized; the nerves and blood vessels are clearly identified; and the fish resumes full opercular function before it is moved to a recovery tank.
The FW environment presents fish with the dual challenge of balancing passive ion losses and osmotic water gain14. The balancing of passive ion losses occurs via the active uptake of salt across the ICs which are localized to the filamental and lamellar epithelia where they can make direct contact with the external environment2,4,8,9,16. However, the location of the ICs on the gill is not static. Over the past three decades a number of studies have shown that several FW fish species, when faced with an ionic and/or temperature challenge, redistribute branchial ICs from the filament or base of the lamella to the more distal regions of the lamella1,2,4,5,17-21. Such redistribution may increase the thickness of the lamellae which can compromise gas transfer (O2, CO2) across the gill epithelia22. Researchers have used the time differential staining technique described in this manuscript to track the relocation and emergence of new ICs on the gill epithelia under these different experimental conditions (Figure 1C)1,4,5.
The gills, and presumably the branchial ICs, are innervated by the IX and X cranial nerves7,23-25. These nerves carry both efferent and afferent inputs to and from the gill, respectively. They are located at the dorsal side of the buccal cavity behind the 4th gill arch. The accessibility of the nerves and the ease with which the bilateral denervation procedure can be performed is species specific. In trout, for example, the pointed and flattened anatomy of the head allows for the nerves to lie in a single plane behind the 4th gill arch under a thin layer of tissue. This makes the nerves visible and easily accessible to the researcher to perform the denervation procedure. In contrast, goldfish have a shorter snout and a rounder head. The IX and X cranial nerves of goldfish lie deeper into the dorsal side of the cavity after the 4th gill arch occupying different planes. This orientation limits the ease of access to the nerves and requires a more careful approach to identify and sever the appropriate nerves. The aim of the denervation procedure is to remove sensory afferent and efferent input to and from the gill, respectively. Denervation of the gill arches can also be coupled with ion flux experiments using radioisotopes (e.g., 22Na). These techniques can be used in tandem to study the contribution of nervous input on ion movement across the gill epithelia. Another limitation to the denervation procedure is the inability to distinguish between sensory and motor neurons, thus when severing the nerve bundle it is possible that both types of innervation are being removed. Severing any motor neurons may affect the gill and opercular movement of the fish. Thus when performing gill denervation experiments it is important to also monitor ventilation after the fish has recovered from the procedure to ensure that there is sufficient gill movement for both gas and ion exchange.
The protocols described in this manuscript use adult animals kept on a 12:12 light:dark cycle and fed commercial food pellets. These methods can be modified in several ways. First, the recovery period after mitochondrion-rich dye exposure can be adjusted to the requirements of the researcher’s protocol (e.g., 1, 3, 5, or 14 days). The longest period of recovery after mitochondrion-rich dye exposure in the lab has been 14 days4,5. There wasn’t a significant decrease in the intensity of the fluorescence of the mitochondrion-rich dye after 14 days of recovery. Second, the use of the α-5 primary antibody is limited to only identifying NKA-rich cells and does not distinguish between the different subtypes of branchial ICs. Fortunately, in goldfish it was established that the majority of ICs are both mitochondrion-rich (label with MitoTracker) and NKA-rich (label with α-5) cells which may not be the case in all fish species4,27. Future experiments can focus on following the temporal redistribution of specific IC subtypes by using antibodies directed specifically against a variety of channels, pumps and exchangers (e.g., NHE, H+ pump). Previous studies found that the majority of ionocytes stained with mitochondrion-rich dye exhibit NKA immunoreactivity4. Innervation of the ICs can be detected by using a primary antibody against zebrafish-derived neuron-specific antigen (zn-12). In this study, α-5 and zn-12 primary antibodies were detected by using the same secondary antibody (Alexa Fluor 488) for both. This limitation is due to both primary antibodies being raised in a mouse host and is overcome by the fact that the NKA-rich cells and neurons can be distinguished morphologically even though they fluoresce the same colour. Sequential staining with different secondary antibodies (e.g., first α-5 with Alexa Fluor 488 then zn-12 with Alexa Fluor 594) can also be used to eliminate the problem of having both markers fluorescing the same color. Lastly, the full bilateral nerve sectioning protocol can be modified to target nerves to specific gill arches. For example, selective nerve sectioning can be performed on the first gill arch by gently separating the first and second gill arches to expose the nerves leading to the first gill arch at the dorsal end of the gill basket (Figure 2B-E). The time differential technique can also be applied to study the distribution of ionocytes in larval zebrafish fish as they develop and ion transport transitions from the skin to the gill.
The authors have nothing to disclose.
We would like to thank William Fletcher for animal care at the University of Ottawa. The authors would also like to thank Dr. William Milsom for teaching VT the full bilateral denervation technique at the University of British Columbia. A travel grant for this research was provided by the Faculty of Graduate and Postgraduate Studies at the University of Ottawa. This research is also supported by NSERC PGS-D scholarship to VT and Discovery Grants Program to SFP.
Name of Material/ Equipment | Company | Catalog Number | Comments/Description |
MitoTracker Red | Life Technologies | M-7512 | |
Dimethyl sulfoxide | Sigma Alrdich | D2650 | |
α-5 primary antibody | Develompental Hybridoma Bank | a5 | |
zn12 primary antibody | Develompental Hybridoma Bank | zn12 | |
Alexa Fluor 488 (anti mouse) | Life Technologies | A-11001 | |
Benzocaine | Sigma Alrdich | E1501 | 4-Aminobenzoic acid ethyl ester, Ethyl 4-aminobenzoate |
Vannas Spring Scissors | Fine Science Tools | 15000-10 | |
Standard Pattern Forceps | Fine Science Tools | 11000-12 | straight |
Standard Pattern Forceps | Fine Science Tools | 11001-12 | curved |
Dumont No. 5 forceps | Fine Science Tools | 11252-30 | |
Tissue retractor | Fine Science Tools | 17009-07 | |
Paraformaldehyde | Sigma Alrdich | P6148 | |
Triton X | Sigma Alrdich | X100 | |
Vectashield with DAPI | Vector Laboratories | H-1200 |