Quantification of Cutaneous Ionocytes in Small Aquatic Organisms

引用 收藏 提问与回复 分享您的反馈 Cited by



Journal of Comparative Physiology B
Feb 2019



Aquatic organisms have specialized cells called ionocytes that regulate the ionic composition, osmolarity, and acid/base status of internal fluids. In small aquatic organisms such as fishes in their early life stages, ionocytes are typically found on the cutaneous surface and their abundance can change to help cope with various metabolic and environmental factors. Ionocytes profusely express ATPase enzymes, most notably Na+/K+ ATPase, which can be identified by immunohistochemistry. However, quantification of cutaneous ionocytes is not trivial due to the limited camera’s focal plane and the microscope’s field-of-view. This protocol describes a technique to consistently and reliably identify, image, and measure the relative surface area covered by cutaneous ionocytes through software-mediated focus-stacking and photo-stitching–thereby allowing the quantification of cutaneous ionocyte area as a proxy for ion transporting capacity across the skin. Because ionocytes are essential for regulating ionic composition, osmolarity, and acid/base status of internal fluids, this technique is useful for studying physiological mechanisms used by fish larvae and other small aquatic organisms during development and in response to environmental stress.

Keywords: Chloride cell (氯离子细胞), Mitochondrion-rich cell (富含线粒体的细胞), Skin (皮肤), Larvae (幼虫), Osmoregulation (渗透调节), pH regulation (pH调节), ATPase (ATP酶), Ocean acidification (海洋酸化)


Ionocytes (formerly called chloride cells and mitochondrion-rich cells) are specialized cells that transport ions to regulate the ionic composition, osmolarity, and acid/base status of internal fluids (reviewed in Evans et al., 2005). Ionocytes typically express high levels of the enzyme Na+/K+ ATPase, which can be used to visualize them using immunohistochemical techniques. In larval and juvenile fishes, ionocytes are present throughout the skin (reviewed in Varsamos et al., 2005). The abundance, distribution, and size of ionocytes can change during development, as well as in response to environmental stressors such as temperature, acidification, salinity, hypoxia, UV light, and pollutants. However, accurate and consistent quantification of cutaneous ionocytes can be difficult due to their presence in large numbers (tens of thousands), non-homogenous distribution, and being masked by pigmented cells. In addition, ionocyte quantification throughout the entire skin of an organism is often limited by the microscope’s field-of-view (Fridman et al., 2011) and both the camera’s field-of-view and focal plane (Hiroi et al., 1999; Varsamos et al., 2002). Because ionocyte distribution is often quite variable across the larval skin, estimations of ionocyte abundance based on a single photograph and focal plane may not be representative of the organism as a whole, and therefore not reliable or repeatable. As such, previous studies that utilized immunohistochemistry to identify and quantify ionocytes have been descriptive in nature (Hwang, 1989; Hiroi et al., 1998; Katoh et al., 2000), which precludes replication and comparison across species and treatments.

Here, we describe a protocol involving immunohistochemistry, photo-microscopy, and photo-editing software that allows estimating ionocyte number, size, density, and relative coverage of cutaneous ionocytes with high accuracy. Unlike previous studies, our method resolves both frame-of-view and focal plane limitations through focus-stacking and photo-stitching software, effectively digitizing a 3-D fish larva into a 2-D image. Additionally, the resulting digital image retains its high-resolution–thereby allowing the accurate quantification of all ionocytes on one side of the larva along with their average size and the larva’s surface area. These parameters can then be inputted into a simple equation to estimate the cutaneous ionocyte ion-transporting capacity in relation to the total skin surface area of the specimen. For larger specimens in which counting all cutaneous ionocytes is cumbersome and time-consuming, we developed a random sampling protocol that allows highly accurate cutaneous ionocyte estimations after counting ionocytes within 10% of the skin surface area. For example, the estimated number of cutaneous ionocytes in a 13.7 mm long Yellowfin Tuna (Thunnus albacares) specimen was 15,342, which was only 3.6% higher than the actual 14,795 cutaneous ionocytes counted in the entire fish skin (Kwan et al., 2019). We believe this method will be valuable for future studies quantifying cutaneous ionocytes in larval and juvenile fish as well as other small aquatic organisms during development and in response to various environmental stress (e.g., ocean warming, hypoxia, and acidification).

Materials and Reagents

  1. PCR Reaction Strips (8 x 0.2 ml) attached flat cap (Simport, catalog number: T320-2N)
  2. 1.7 ml microcentrifuge tubes (Corning, catalog number: MCT-175-C)
  3. 15 ml plastic tubes (Corning Brand, catalog number: 352196) 
  4. Microscope slides (Thermo Fisher Scientific, catalog number: 12-550-15) or concave microscope slides (United Scientific Supplies, catalog number: CS3X11)
  5. Microscope coverslips (e.g., VWR, catalog number: 48366 045)
  6. Microscope small Petri dish trays (MatTek Corp, catalog number: P35GC-1.5-10-C)
  7. Petri dish (Corning Brand, catalog number: 351029) or any other container
  8. Specimen of interest
    Note: This protocol is optimized for small (< 30 mm) samples with relatively low natural pigmentation and. high transparency. This protocol could be used for other specimens, but additional optimization of immunostaining and image acquisition might be required. 
  9. Deionized (DI) water
  10. 10x Phosphate buffer saline (PBS) (Corning Incorporated, catalog number: 46-013-CM)
  11. 20% paraformaldehyde (PFA) (Electron Microscopy Sciences, catalog number: 15713)
  12. 200-proof (100%) Ethyl alcohol (Decon Laboratories, catalog number: 2701)
  13. 30% (w/v) hydrogen peroxide (Sigma-Aldrich, catalog number: H3410)
  14. α5 Na+/K+ ATPase (NKA) antibody (Developmental Studies Hybridoma Bank, catalog number: a5–supernatant)
    1. This monoclonal antibody has specifically recognized NKA from multiple classes of aquatic organisms including: bony fishes (Wilson et al., 2000 and 2002; Yang et al., 2013; Tang et al., 2014; Kwan et al., 2019), elasmobranch fishes (Piermarini and Evans, 2001; Roa et al., 2014; Roa and Tresguerres, 2017), and hagfish (Clifford et al., 2015).
    2. This protocol may be adapted for other proteins following validation of primary antibodies in the species of interest. However, successful localization using different primary antibodies may require optimizing the fixation and immunohistochemical procedures described here.
  15. Vectastain® Elite® ABC HRP ready-to-use kit (Vector Laboratories, catalog number: PK-7200), includes:
    1. Normal horse serum blocking solution
    2. Pan-specific secondary antibody solution 
    3. Streptavidin peroxidase solution
  16. 3,3-diaminobenzidine (DAB) Peroxidase Substrate stored at 4 °C (Vector Laboratories, catalog number: SK-4100)
  17. Sodium azide (Thermo Fisher Scientific, catalog number: S227I-25)
  18. 1.2x phosphate buffer saline (PBS) solution (see Recipes)
  19. 4% paraformaldehyde (PFA) fixative solution (see Recipes)
  20. 3% hydrogen peroxide solution (see Recipes)
  21. 1.0x PBS solution (see Recipes)
  22. DAB substrate working solution (see Recipes)
  23. 10% sodium azide stock solution (see Recipes)
  24. 0.02% sodium azide working solution (see Recipes)


  1. Fridge (4 °C) and freezer (-80 °C)
  2. Chemical fume hood
  3. Borosilicate glass vial (Thermo Fisher Scientific, catalog number: 033374)
  4. Shake table (e.g., VWR, Standard Orbital Shaker, Model 1000) or rotator (e.g., Thermo Fisher Scientific, Tube Revolver/Rotator, catalog number: 88881001)
  5. Transfer pipette (Thermo Fisher Scientific, catalog number: 1368050) or any other pipette suitable for your specimen of interest
  6. Fine forceps (Fine Science Tools, catalog number: 26029-10) or any forceps that does not damage your specimen of interest
  7. Camera-mounted microscope(s) Leica DMR compound microscope for smaller samples (< 5 mm SL), Leica MZ9.5 stereomicroscope for larger samples (> 5 mm)
  8. Microscope light source (Ludl Electronic Products [LEP] [12 V, 100 W], catalog number: 99019)
  9. Microscope mounted camera (Canon, Rebel T3i single lens reflex camera)
  10. Compatible camera with SD card
  11. Anti-vibration table for microscopy imaging
  12. Microscope stage micrometer standard
  13. Monitor for microscope camera’s live-feed imaging
  14. Camera remote for wireless shutter release (Canon RC-6)


  1. Helicon Focus (HeliconSoft, Version 6.7.2 or later, www.heliconsoft.com)
  2. Adobe Photoshop (Adobe Systems, version CS3 or higher, www.adobe.com)
  3. Fiji (https://imagej.net/Fiji), or ImageJ (https://imagej.nih.gov/ij/download.html) with cell counter plugin (https://imagej.nih.gov/ij/plugins/cell-counter.html)
  4. Adobe Illustrator (Adobe Systems, Version 19.2.0 or later, www.adobe.com)
  5. R (https://www.r-project.org)


An overview of this protocol is shown below (Figure 1). The time necessary to finish this procedure varies with the size of each sample. At the minimum, this procedure will take at least 3 days.

Figure 1. Procedural Overview. Schematic showing the major steps of this protocol.

  1. Sample fixation
    1. Euthanize specimen in accordance with an approved animal care protocol.
    2. Incubate samples in fixative solution (see Recipes 1 and 2) in any water-tight container (e.g., borosilicate glass vial) on a rotator/shake table overnight (8-12 h) at 4 °C.
      1. All samples should be fully immersed in the fixative. Multiple samples can be fixed together. In Kwan et al., 2019, 6-10 fish larvae (ranging from 3-25 mm in length) were fixed in 10 ml of fixative within a 20 ml borosilicate glass vial. Adjust container, volume, and fixative according to your specimen of interest. Rinsing the samples prior to fixation is not required.
      2. The ethanol washes (Steps A3 and A4) can be replaced with three 10 min PBS washes if the samples are to be immediately processed for immunohistochemistry.
    3. Remove fixative, then incubate samples in 50% ethanol on a shake table overnight (8-12 h) at 4 °C.
    4. Remove 50% ethanol, then incubate samples in 70% ethanol for long-term storage at 4 °C.

  2. Whole-mount immunohistochemistry
    1. Begin rehydrating samples by transferring onto a Petri dish with tap water for 1 min.
    2. After initial rinse, transfer samples into a plastic tube (e.g., 0.2 ml PCR tube or 1.7 ml microcentrifuge tube depending on larval size) with tap water (filled to max volume) and further incubate for 10 min at room temperature on a rotator/shake table.
      1. All samples should be fully immersed in solution. In Kwan et al., 2019, fish larva (< 20 mm in length) were incubated in a 1.7 ml microcentrifuge tube and immersed in ~1 ml of solution. Larger samples will require a larger container and more solution (and vice versa). Adjust container and solution volume accordingly.
      2. During solution changes, it is preferable to remove the solution from the container–as opposed to moving the sample to containers with different solutions. This minimizes damage to the sample. This applies to all solution changes.
      3. If changing solution with a plastic transfer pipette becomes problematic, consider using a more appropriately sized pipette (P200 pipette with a 200 μl tip for 0.2 ml PCR tube and P1000 pipette with a 1,000 μl tip) for changing solutions. Adjust according to your sample and/or container.
    3. Remove water, then incubate samples in freshly prepared 3% hydrogen peroxide (see Recipe 3) for 10 min at room temperature on a rotator/shake table.
      1. During solution changes, it is preferable to remove the solution from the container–as opposed to moving the sample to containers with different solutions. This minimizes damage to the sample. This applies to all solution changes.
      2. This step may be extended to 1 h if there is intense pigmentation and/or non-specific background staining.
    4. Remove 3% hydrogen peroxide solution, then incubate samples in normal horse serum (NHS) blocking solution for 15 min at room temperature on a rotator/shake table.
    5. Prepare primary antibody solution (e.g., α5 NKA antibody: 20 ng/ml in NHS blocking solution).
    6. Remove NHS, then incubate samples in primary antibody solution and leave on a rotator/shake table overnight (≥ 8 h) at room temperature or (preferably) at 4 °C.
      Note: This step may be modified for various antibody validation controls. See Procedure K for details.
    7. Remove primary antibody solution, then rinse samples in 1x PBS (see Recipe 4) for 5 min at room temperature on a rotator/shake table. Repeat two more times for a total of three rinses.
    8. Remove 1x PBS wash, then incubate samples in pan-specific secondary antibody solution for 30 min at room temperature on a rotator/shake table.
    9. Remove pan-specific secondary antibody solution, then rinse samples in 1x PBS for 5 min at room temperature on a rotator/shake table. Repeat two more times for a total of three rinses.
    10. Remove 1x PBS wash, then incubate samples in streptavidin peroxidase solution for 15 min at room temperature on a rotator/shake table.
    11. Remove streptavidin peroxidase solution, then rinse samples in 1x PBS for 5 min at room temperature on a rotator/shake table. Repeat two more times for a total of three rinses.
    12. Prepare the DAB substrate working solution (see Recipe 5) according to the manufacturer’s instruction in a separate container (e.g., 15 ml centrifuge tube).
    13. Remove 1x PBS wash, then incubate samples in DAB substrate working solution for 20 s at room temperature.
      1. The sensitivity of this portion is highly dependent on the tissue sample and antibody of interest. Because DAB staining is irreversible, start with 20 s and increase incubation time if darker staining is necessary. Optimal incubation time should be determined by the investigator(s).
      2. Incubating samples for a longer duration may result in darker background staining. If this becomes an issue, consider diluting the DAB substrate working solution with DI water.
    14. Remove DAB substrate working solution, then rinse samples in DI water for 10 min at room temperature on a rotator/shake table. Repeat two more times for a total of three rinses.
    15. Store in DI water at 4 °C until imaging.
    16. For long-term storage (1 week or longer), store in PBS + 0.02% sodium azide (see Recipes 6 and 7).
      Note: Sodium azide is very toxic. Samples incubated within sodium azide should be given multiple DI water washes before handling and imaging.

  3. Light microscopy and imaging
    1. Carefully transfer sample onto a microscope slide, concave microscope slide, or a glass bottom dish for examination under a light microscope.
    2. Add DI water to keep sample from drying out, and secure sample with a coverslip to prevent unintentional movement.
    3. Switch on the live feed option on the DSLR camera.
    4. Switch off the autofocus option on the DSLR camera.
    5. Set camera to manual or aperture priority mode. Adjust the following parameters until image is optimal: camera ISO, camera aperture, microscope aperture, light source brightness.
    6. For optimal imaging, use a wireless remote controller to trigger shutter release.
      Note: Avoid bumping into the entire camera apparatus. Because the exposure time of each photograph will likely require ~1 s per image, any movement of the microscope, sample, camera, and surrounding can cause blurriness.
    7. Starting at one end (e.g., head of fish specimen), photograph sample at each focal plane (Z-stack) within the field of view. Manually adjust the focus until the entire Z-stack is finished (Figures 2A-2E). Be careful to not move the image in the X or Y-dimension.
      1. The image may differ between the microscope lens and camera (shown on the live feed monitor). Always use the camera’s focal plane and image accordingly.
      2. Shooting the Z-stack in consecutive order (front to back or vice versa) is preferable as it may help during post-processing.
      3. Ensure that your sample does not drift in the x or y plane between images for proper post-processing.
      4. An exact Z-step (distance between each image) ranges greatly amongst samples. After each image, turn the fine focus until the next plane is in focus. In Kwan et al., 2019, most images took around 6-8 images per Z-stack.

      Figure 2. Example of Z-stacks and focus stacking. After photographing a Z-stack on the ventral body (A-E), the focused portions of each image were stacked together via focus stacking (F).

    8. Move to the next unphotographed area, taking care to overlap consecutive Z-stacks by at least ~20% to ensure photo stitching accuracy (Figure 3).

      Figure 3. Overlap of Z-stacks are necessary for proper photo-stitching. Each Z-stack (A, B) should have ~20% of overlapping area for optimal photo-stitching (C).

    9. Continue imaging Z-stacks until the entire sample has been photographed.
    10. Photograph the microscope standard at a height equal to approximately half of the sample’s thickness.
      Note: The same scale bar can be applied across multiple samples as long as the magnification of the microscope has not changed and the sample thickness is relatively similar.

  4. Focus-stacking and photo-stitching
    1. A video tutorial explaining the focus-stacking process can be found at youtube–Quick Start with Helicon Focus (published by HeliconSoft).
    2. A video tutorial explaining the photo-stitching process can be found at youtube–How to Stitch Multiple Photos Together Automatically with Photomerge in Photoshop (published by Photoshop Video Academy).
    1. Import each stack of images into Helicon Soft and render focus-stacking.
      1. Begin with Method C (pyramid). 
      2. If the edge of the image becomes distorted (halo effect), troubleshoot with Method A (weighted average) or Method B (depth map)–and increase the “radius” value. Read more about the different methods here: https://www.heliconsoft.com/helicon-focus-main-parameters/.
      3. Other focus-stacking software (e.g., Adobe Photoshop) may be used, though the processing speed and quality of results may vary.
    2. Repeat for each stack of images. Save images.
    3. Open all focus-stacked images in Adobe Photoshop and check for processing artifacts (Figure 4). If artifacts exist, use the “Crop” tool (hotkey: C) to discard the artifact. Save all changes before proceeding.

      Figure 4. Example of artifacts created during focal stacking. Artifacts such as blurred areas along the edge of the Z-stack may appear (black arrow) and should be cropped out of the image.

    4. With all focus-stacked images opened in Adobe Photoshop and vetted for artifacts, use the photo-stitching tool (File > Automate > Photomerge …).
    5. Select “Add Open Files” (all files must be already saved), select “Auto”, then “OK”.
      Note: If there are many stacks to stitch together, this step may take a while. If computer crashes, try using a computer with more RAM and stronger processor. 
    6. Check for artifacts or unphotographed areas. Save image.
    7. Open the image containing microscope standard taken back in Step C10. Using the “Rectangle Marquee” tool (hotkey: M) to measure a known width (e.g., 1 mm). 
    8. Use the “Brush” tool (hotkey: B) to fill in the rectangle with a color that contrasts well with the background.
    9. With the colored rectangle still in selection, copy (Windows hotkey: Ctrl + C; Mac hotkey: Cmd + C) the scale bar and paste (Windows hotkey: Ctrl + V; Mac hotkey: Cmd + V) onto the photo-stitched image of the sample. Save image.

  5. Calibrating measurement tools for image analysis/ionocyte quantification
    Note: A video tutorial explaining this process can be found at Youtube–Set calibration in ImageJ (published by remotelab PloyU).
    1. Open the photo-stitched image in FIJI/ImageJ.
    2. Using the “magnifying glass” tool or by pushing the “+” key, zoom in on the scale bar.
      Note: Hover the mouse cursor over the area you want to zoom in before pushing “+”. You can also push “-” to zoom out.
    3. Using the “straight line” tool, draw a line across the scale bar–holding shift to lock the line at a 90° angle. 
    4. At the top of the window (screen for Mac), click on Analyze > Set Scale. The distance in pixels should already be filled in as that is the line you just drew. Fill in the “Known Distance” and “Unit of Length” (e.g., 1 mm), then click “OK”.
      Note: If the same scale bar is being used across multiple samples, then select the box next to “Global”. In doing so, the pixel to mm ratio will be remembered across windows (samples).
    5. To ensure accuracy, open up the microscope standard image previously taken during “Step C10”. Using the “straight line” tool, draw a line across the imaged scale bar and measure (hotkey: M). If done properly, the difference between the imaged scale bar and the “measured” distance should be negligible.

    Note: From this point forward, this protocol will regard the sample as a fish specimen. Other samples could be similarly quantified, though some modifications to this protocol may be necessary.

  6. Surface area estimation
    1. Using FIJI/ImageJ’s “straight line” tool, measure (hotkey: M) the standard length of the fish larva. Record this value.
      Note: The “segmented line” tool (right click (Mac: command + click) on the “straight line” tool) can be used if the fish larva was fixed or placed in a curved position.
    2. With the “Freehand Selection” tool, trace the fish larva to measure (hotkey: M) its surface area. Record this value.
      Note: If freehand drawing using a mouse proves difficult, try using a tablet and stylus.
    3. Using the “Freehand Selection” tool, trace regions of interest where ionocyte distribution may differ (in this example we use head, body, and fins) and measure (hotkey: M). Record these values.

  7. Counting cutaneous ionocytes
    1. This technique is suitable for smaller fish specimens (< 5 mm standard length) and samples with a non-homogenous distribution of ionocytes.
    2. A video tutorial explaining this process can be found at Youtube–Count Items in ImageJ (published by Timothy Spier).
    1. Open FIJI/ImageJ.
    2. If using ImageJ, be sure the “Cell Counter” plugin has been installed (see Software).
    3. Open the photo-stitched image in FIJI/ImageJ.
    4. Go to Plugin > Analyze > Cell Counter > Cell Counter.
    5. In the Cell Counter window, click “Initialize”, select “Type 1”, and zoom (“+”) in on the area of interest. 
    6. Using your mouse, left click to denote a cell at that location. The count will be updated next to “Type 1” in the Cell Counter window. Record this value.
    7. To remove an erroneous count, check the box next to “Delete Mode”. Next, click on the point you would like to remove. Uncheck the box to proceed with counting.
    8. Use the “Save Markers” option in the Cell Counter window to save your work.

  8. Estimating number of cutaneous ionocytes
    Note: This technique is suitable for larger fish specimens with a homogenous ionocyte distribution. Because there is likely species-specific variation in ionocyte distribution, this estimation technique needs to be ground truthed with manual cell counts (Procedure G) before utilizing.
    1. Open the photo-stitched image with scale bar in Adobe Illustrator.
      1. To zoom in, use the zoom tool (hotkey: Z) and click.
      2. To zoom out, use the zoom tool (hotkey: Z), hold ctrl (Mac: hold Cmd) and click.
    2. Adjust the artboard (hotkey: Shift + O) to fit the photo-stitched image. If the photo-stitched image is larger than the maximum artboard size, scale down the image size by 1) clicking on the image, 2) hovering over a corner anchor point, 3) holding “shift”, and 4) dragging to reduce image size.
    3. Change the name of this layer to “image”, and lock it. Create four additional layers and arrange them in this order: fin count, body count, head count, grid, image (from top to bottom) (Figure 5).
      1. You may have other areas of interest requiring additional layers
      2. A layer is only editable when it is 1) visible, 2) unlocked, and 3) selected.
      3. Layer(s) at the top will cover the layer(s) at the bottom.

      Figure 5. Key buttons within the layers window. The eye symbol (A) controls the visibility of that layer, whereas the lock symbol (B) controls whether that layer is editable. To change the name of the layer, double click on the name (C). To select everything in that layer, click on the circle in the layer of interest (D). To add another layer, click “create new layer” (E).

    4. Toggle the “Ruler” option (Window: Ctrl + R; Mac: Cmd + R). Right click on the ruler and change to “pixels”.
    5. Using the rectangle tool (hotkey: M), drag a rectangle across the scale bar (Figure 6A).
    6. On the second row from the top towards the right side of the window, find and click the orange “Transform”. 
    7. Record the value listed next to “W” (Figure 6B). This value will assist you in converting between pixels and the measured metric unit.

      Figure 6. Finding the conversion between digital pixels and measured metric units. Draw a rectangle across the scale bar (A). The width of the rectangle should then appear in the “Transform” window (B). Use this value to convert between the computer pixels and the metric system.

    8. Using the standard length that was calculated back in Step F1, calculate the value equal to 2% of the fish specimen’s standard length.
    9. Using the rectangle tool, click (not drag) to create a shape with custom width and height dimensions. Input the 2% standard length value into both the “Width” and “Height” dimension (make a square).
      Note: For example, if the standard length of specimen = 8.80 mm, then 2% of the standard length is 0.176 mm.
    10. Copy and paste the squares until a grid overlays the entire fish (Figure 7).
      Note: To expedite this process, below are some helpful shortcuts:
      1. Select All: “Windows: Ctrl + A, Mac: Cmd + A”
      2. Copy: “Windows: Ctrl + C; Mac: Cmd + C”
      3. Paste in place: “Windows: Ctrl + Shift + V; Mac: Cmd + Shift + V”
      4. While dragging newly pasted cells, hold “Shift” to lock the squares in the X or Y plane.
      5. By default, illustrator should “auto-snap” your selection to place. If this is turned off, go to Preferences > Selection and Anchor Display and check the box “Snap to Point”.

      Figure 7. Example of an overlying grid over a photo-stitched yellowfin tuna specimen. Each square’s length is equal to 2% of the larva’s standard length.

    11. Number the grid’s x and y axes.
    12. Calculate the area of an individual square. Record this value.
      Note: For example, if the standard length of specimen = 8.80 mm, then the area of each square is 0.031 mm2.
    13. Using the measurements from Steps F2 and F3, calculate how many squares will require counting to cover 10% of each region of interest (e.g., head, body, fins). Record the number of squares necessary to cover at least 10% of each region.
      1. If the area of the head region = 4.9 mm2 and each square is 0.031 mm2, then: . Round up and count 16 squares to adequately cover at least 10% of the head region.
      2. Separation of different body regions is necessary as variation in regional ionocyte density may occur due to species-specific or developmental stage-specific differences.
    14. Open R. Copy and paste in the provided codes (see Note #1 below). Adjust the number for the X and Y coordinates accordingly before using.
      Note: Step-by-step instruction for using this R below:
      ### Note #1
      ### Beginning of code
      # Generating random coordinates
      # Step-by-step instruction
      # Objective: this will randomly generate 500 (x,y) coordinates
      # Step 1) In line 15, replace "59" with the number of squares on the x-axis
      # Step 2) In line 16, replace "17" with the number of squares on the y-axis
      # Step 3) highlight everything (Windows: ctrl + A; Mac: cmd + A)
      # Step 4) hold "Shift" and hit "enter"
      # Step 5) see output in the R Console window.
      # Note 1: Some coordinates may be repeated; skip if repeated.
      # Note 2: If 500 (x,y) coordinates is not enough, repeat until the quota for each region is reached.

      x.coord = sample(1:59, replace=T, 500)
      y.coord = sample(1:17, replace=T, 500)
      rand.coords <- data.frame(x.coord,y.coord)

      ### End of code

    15. For each randomly generated coordinate, disqualify for counting if the square contains any of the following:
      1. Background space, glare, or other photography artifacts (Figure 8A).
      2. Folding, invaginations, pigments, which prevent accurate counts (Figure 8B).
      3. More than one region of interest: head, body, fins (Figure 8C).

        Figure 8. Example of squares unsuitable for cell counting. Randomly generated coordinates must not contain the following features: background space (A), foldings, invaginations, pigments, or other dark space (B), and space inclusive of more than one region e.g., box containing both body and fin (C).

    16. If the square meets the above criteria, then mark for counting.
      1. To quickly designate the square for counting, use the “Pencil tool” (hotkey: N) to scribble a marking.
      2. To avoid accidently moving the grid or the underlying image, be sure to lock both the grid and image layer (see Figure 5B).
    17. Count the cells within Adobe illustrator. For each region, record the number of ionocytes present in the randomly-chosen grid squares.
      Note: To quickly record the number of cells in the square, use the “Type tool” (hotkey: T) and click within the square, then type the number of counted cells. Hit “Esc” to exit this mode.
    18. If the grid lines happen to overlap the immunostained cells, count only the cells intersecting with the top and right sides of the square. This will ensure no double-counting of the cells if two squares were randomly chosen next to each other.
      Note: If the grid lines are too thick or the cells are located directly underneath the lines, try selecting the grid layer (hit the rightmost circle in the grid layer; Figure 5D) and reducing the “stroke” value to “0.5”, or reducing the opacity to 50% for better viewing.
    19. Continue until at least 10% of each region (head, body, fins) have been counted (Figure 9).

      Figure 9. Method for quantification of cutaneous ionocytes in yellowfin tuna specimen and early stage juveniles (> 5 mm standard length). Ionocytes were identified by their intense Na+/K+-ATPase immunostaining and counted within randomly sampled boxes of the overlaid grid within the head (green), body (pink), and fin (blue) regions. Dashed white lines outline regions that were not sampled due to heavy pigmentation preventing accurate ionocyte counts. The estimated ionocyte count (15,342 cells) was ground truthed with manual ionocyte count (colorless regions; 14,795 cells). The estimated relative ionocyte area for this larva is 3.75%. Figure from Kwan et al., 2019.

    20. Calculate the estimation using the following equation:

      1. Areahead, body, or fins = measured area of head, body, or fins region (Step F3).
      2. Ionocytehead, body, or fins = sum of ionocytes counted per region (Step H17).
      3. Squarehead, body, or fins = number of squares counted per region (Step H13).
      4. Areasquare = size of each square = (standard length*0.02)2 (Step H12).

  9. Quantifying ionocyte size
    1. Open up ImageJ and set scale bar as discussed in Procedure E.
    2. With the “Freehand Selection” tool, zoom in (hotkey: +) and trace individual cutaneous ionocytes to estimate their relative surface areas.
      Note: If freehand drawing proves difficult, try using a tablet and stylus.
    3. Repeat Step I2 until you have measured 20 cells from different regions of interest (e.g., head, body, and fins).
    4. Calculate the average ionocyte size using the following equation:

      Areahead, body, or fins were calculated in Step F3.
      Areatotal was calculated in Step F2.

  10. Quantifying relative ionocyte area

    Estimated total ionocytes were calculated in Step H20.
    Average ionocyte size were calculated in Step I4.
    Total surface area were calculated in Step F2.

  11. Method controls and antibody validation
    1. Non-specific secondary and tertiary antibody binding control: designate several fish specimens as negative controls by omitting the primary antibody during Step B6 (leave in NHS blocking solution overnight at 4 °C). Fish specimens incubated with the primary NKA antibody should result in positive staining (Figure 10A). In contrast, fish specimens without the primary antibody should result in the absence of staining (Figure 10B).
      1. Performing this negative control step is a great way to identify and avoid the natural pigments present on the organism of interest (Step H15).
      2. If necessary, the negative-control fish specimens can be re-incubated and utilized. Simply repeat all of Procedure B to re-immunostain said sample(s).

      Figure 10. Negative control for immunohistochemistry. The difference between a yellowfin tuna specimen incubated with (A) and without (B) Na+/K+-ATPase (NKA) antibodies. The brown precipitate seen in the top larva denotes NKA presence on the cutaneous surface, whereas the absence of brown staining on the bottom larva serves a successful negative control. Note the areas containing naturally occurring pigment in both specimens should be avoided during estimation.

    2. Primary antibody specificity via western blotting: flash frozen tissue (-80 °C) should be sampled along with the fixed tissue for western blotting, which should yield a single band close the predicted size of your protein of interest (Figure 11). See supplementary material 2 in Kwan et al. (2019) for protocol and additional details.

      Figure 11. Western blot of Na+/K+-ATPase (NKA) in yellowfin tuna tissue. Western blot with anti-NKA monoclonal antibodies on yellowfin tuna tissue yielded a single ~108 kDa band, which matched the predicted size of the protein. Figure adapted from Kwan et al., 2019.

Data analysis

Ionocyte count (Step H20), average ionocyte size (Step I4), and relative ionocyte area (Step J1) data across development is best depicted in a scatterplot and explained using regression models. In Kwan et al., 2019, we examined the relationships of relative ionocyte area across standard length and days post hatching using linear regressions. The slope of the linear regression explains the steepness of the line, with the null hypothesis being zero, whereas R2 describes the regression’s goodness-of-fit. Other developmental metrics that may be applicable include accumulated thermal units and total length of the sample. Development may not always be linear; therefore, researchers should choose their own regressions according to your field of study. If the experiment includes multiple treatments or species, the comparison of the regression’s slope should be examined with t-test (2 treatments total) or Analysis of Covariance (3 or more treatments). See Zar (2014) for additional details.


  1. 1.2x phosphate buffer saline (PBS) solution (1 L)
    120 ml of 10x PBS stock solution
    880 ml of deionized water
  2. 4% paraformaldehyde (PFA) fixative solution (10 ml)
    a. Break open a PFA ampule (20% stock solution) within the fume hood
    b. Add 2 ml of PFA stock solution into 8 ml of 1.2x PBS solution (Recipe 1), for a final concentration of 4% PFA in 1 M PBS. Invert to mix
    c. Remaining PFA stock solution can be stored at 4 °C for up to six months
    d. 4% PFA fixative solution can be stored at 4 °C for up to one week
  3. 3% hydrogen peroxide solution (10 ml)
    1 ml of 30% hydrogen peroxide
    9 ml of tap water
  4. 1.0x PBS solution (1 L)
    100 ml of 10x PBS stock solution
    900 ml of deionized water
  5. DAB substrate working solution (same as manufacturer instruction) (~5 ml)
    5 ml of DI water
    2 drops of buffer stock solution (DAB substrate kit)
    4 drops of DAB stock solution (DAB substrate kit)
    2 drops of hydrogen peroxide solution (DAB substrate kit)
  6. 10% sodium azide stock solution (10 ml)
    10 ml of 1.0x PBS
    1 g of sodium azide
  7. 0.02% sodium azide working solution (10 ml)
    10 ml 1.0x PBS
    20 μl of 10% sodium azide stock solution (Recipe 6)


This protocol corresponds to the methods and experiments presented in Kwan et al. (2019). The larval yellowfin tuna and white seabass samples (Atractoscion nobilis) used to create and optimize this protocol were provided by the Inter-American Tropical Tuna Commission (IATTC) and the Hubbs Sea World Research Institute (HSWRI). We are grateful to Jeanne Wexler (IATTC) for her contributions to the original project, and for rearing and sampling the yellowfin tuna larvae. We thank Diomedes Ballesteros, Lina Castillo, Susana Cusatti, Danisin Dominguez, Agustin Ortega, Daniel Perez, Dario Ramires, Daniel Solis, Luis Tejada, and Carlos Vergara for assisting with the production and rearing of yellowfin tuna larvae at the Achotines Laboratory in Panama. We thank Daniel Margulies (IATTC) and Vernon Scholey (IATTC) for developing the yellowfin tuna rearing methods and supervising the spawning and larval rearing at the Achotines Laboratory. We are grateful to Erica Brombay-Fanning (HSWRI), Eric McIntire (HSWRI), Sabrina Sobel (HSWRI), and Christy Varga (HSWRI) for the rearing and sampling the white seabass larvae of larval white seabass. We thank Mark Drawbridge (HSWRI) for supervising the spawning and larval rearing at the Leon R. Raymond Marine Fish Hatchery in Carlsbad, USA. The authors would like to thank Dr. Greg Rouse for the use of microscope and camera equipment, Sabine Faulhaber for technical assistance with the scanning electron microscope, and Taylor Smith for her assistance in dissection and imaging. G.T.K. was supported by the San Diego Fellowship and the National Science Foundation Graduate Research Fellowship Program.

Competing interests

The authors declare no competing interests.


All experiments were approved by the Scripps Institution of Oceanography/University of California, San Diego animal care committee under protocol no. S10320 in compliance with the Institutional Animal Care and Use Committee (IACUC) guidelines for the care and use of experimental animals.


  1. Clifford, A. M., Goss, G. G., Roa, J. N. and Tresguerres, M. (2015). Acid/base and ionic regulation in hagfish. In: Hagfish Biology. Science Publishers, Boca Raton, pp. 277-292.
  2. Evans, D. H., Piermarini, P. M. and Choe, K. P. (2005). The multifunctional fish gill: dominant site of gas exchange, osmoregulation, acid-base regulation, and excretion of nitrogenous waste. Physiol Rev 85(1): 97-177.
  3. Fridman, S., Bron, J. E. and Rana, K. J. (2011). Ontogenetic changes in location and morphology of chloride cells during early life stages of the Nile tilapia Oreochromis niloticus adapted to fresh and brackish water. J Fish Biol 79(3): 597-614.
  4. Hiroi, J., Kaneko, T., Seikai, T. and Tanaka, M. (1998). Developmental sequence of chloride cells in the body skin and gills of Japanese flounder (Paralichthys olivaceus) larvae. Zoolog Sci 15(4): 455-460.
  5. Hiroi, J., Kaneko, T. and Tanaka, M. (1999). In vivo sequential changes in chloride cell morphology in the yolk-sac membrane of mozambique tilapia (Oreochromis mossambicus) embryos and larvae during seawater adaptation. J Exp Biol 202 Pt 24: 3485-3495.
  6. Hwang, P. P. (1989). Distribution of chloride cells in teleost larvae. J Morphol 200(1): 1-8.
  7. Katoh, F., Shimizu, A., Uchida, K. and Kaneko, T. (2000). Shift of chloride cell distribution during early life stages in seawater-adapted Killifish, Fundulus heteroclitus. Zoolog Sci 17(1): 11-18.
  8. Kwan, G. T., Wexler, J. B., Wegner, N. C. and Tresguerres, M. (2019). Ontogenetic changes in cutaneous and branchial ionocytes and morphology in yellowfin tuna (Thunnus albacares) larvae. J Comp Physiol B 189(1): 81-95.
  9. Piermarini, P. M. and Evans, D. H. (2001). Immunochemical analysis of the vacuolar proton-ATPase B-subunit in the gills of a euryhaline stingray (Dasyatis sabina): effects of salinity and relation to Na+/K+-ATPase. J Exp Biol 204(Pt 19): 3251-3259.
  10. Roa, J. N., Munevar, C. L. and Tresguerres, M. (2014). Feeding induces translocation of vacuolar proton ATPase and pendrin to the membrane of leopard shark (Triakis semifasciata) mitochondrion-rich gill cells. Comp Biochem Physiol A Mol Integr Physiol 174: 29-37.
  11. Roa, J. N. and Tresguerres, M. (2017). Bicarbonate-sensing soluble adenylyl cyclase is present in the cell cytoplasm and nucleus of multiple shark tissues. Physiol Rep 5(2). pii: e13090.
  12. Tang, C. H., Leu, M. Y., Yang, W. K. and Tsai, S. C. (2014). Exploration of the mechanisms of protein quality control and osmoregulation in gills of Chromis viridis in response to reduced salinity. Fish Physiol Biochem 40(5): 1533-1546.
  13. Varsamos, S., Diaz, J. P., Charmantier, G., Blasco, C., Connes, R. and Flik, G. (2002). Location and morphology of chloride cells during the post-embryonic development of the european sea bass, Dicentrarchus labrax. Anat Embryol (Berl) 205(3): 203-213.
  14. Varsamos, S., Nebel, C. and Charmantier, G. (2005). Ontogeny of osmoregulation in postembryonic fish: a review. Comp Biochem Physiol A Mol Integr Physiol 141(4): 401-429.
  15. Wilson, J. M., Randall, D. J., Donowitz, M., Vogl, A. W. and Ip, A. K. (2000). Immunolocalization of ion-transport proteins to branchial epithelium mitochondria-rich cells in the mudskipper (Periophthalmodon schlosseri). J Exp Biol 203(Pt 15): 2297-2310.
  16. Wilson, J. M., Whiteley, N. M. and Randall, D. J. (2002). Ionoregulatory changes in the gill epithelia of coho salmon during seawater acclimation. Physiol Biochem Zool 75(3): 237-249.
  17. Yang, W. K., Kang, C. K., Chang, C. H., Hsu, A. D., Lee, T. H. and Hwang, P. P. (2013). Expression profiles of branchial FXYD proteins in the brackish medaka Oryzias dancena: a potential saltwater fish model for studies of osmoregulation. PLoS One 8(1): e55470.
  18. Zar, J.H. (2014). Comparing simple linear regression. In: Biostatistical Analysis. In: Pearson, H. (Ed.). pp: 387-404.


【摘要】水生生物具有称为离子细胞的特化细胞,其调节内部流体的离子组成,渗透压和酸/碱状态。在小型水生生物如早期生命阶段的鱼类中,通常在皮肤表面上发现离子细胞,并且它们的丰度可以改变以帮助应对各种代谢和环境因素。离子细胞大量表达ATP酶,最显着的是Na + / K + ATP酶,可以通过免疫组织化学鉴定。然而,由于相机的焦平面和显微镜的视野有限,皮肤离子细胞的量化并不是微不足道的。该协议描述了一种技术,通过软件介导的聚焦堆积和照片缝合,一致且可靠地识别,成像和测量皮肤离子细胞覆盖的相对表面积,从而允许皮肤离子细胞面积的量化作为离子传输能力的代表。穿过皮肤。由于离子细胞对于调节内部流体的离子组成,渗透压和酸/碱状态至关重要,因此该技术可用于研究鱼类幼虫和其他小型水生生物在发育过程中和环境胁迫下所使用的生理机制。

【背景】离子细胞(以前称为氯离子细胞和富含线粒体的细胞)是运输离子以调节内部液体的离子组成,渗透压和酸/碱状态的特化细胞(Evans et al。>,2005年综述) )。离子细胞通常表达高水平的Na + / K + ATP酶,其可用于使用免疫组织化学技术使它们可视化。在幼虫和幼鱼中,离子细胞存在于整个皮肤中(综述于Varsamos 等人,>,2005)。离子细胞的丰度,分布和大小可以在发育过程中发生变化,也可以响应环境压力因素,如温度,酸化,盐度,缺氧,紫外线和污染物。然而,皮肤离子细胞的准确和一致的定量可能是困难的,因为它们存在大量(数万),非均匀分布,并被色素细胞掩盖。此外,生物体整个皮肤的离子细胞定量通常受到显微镜视野的限制(Fridman et al。>,2011)以及相机的视野和焦平面(Hiroi et al。>,1999; Varsamos et al。>,2002)。因为离子细胞分布在幼虫皮肤上通常是可变的,基于单个照片和焦平面的离子细胞丰度的估计可能不能代表整个生物体,因此不可靠或不可重复。因此,以前使用免疫组织化学鉴定和定量离子细胞的研究本质上是描述性的(Hwang,1989; Hiroi et al。>,1998; Katoh et al。>,2000 ),这排除了物种和治疗之间的复制和比较。

在这里,我们描述了一个涉及免疫组织化学,光学显微镜和照片编辑软件的协议,可以高精度地估计皮肤离子细胞的离子细胞数量,大小,密度和相对覆盖率。与以前的研究不同,我们的方法通过聚焦堆叠和照片拼接软件解决了视野和焦平面限制,有效地将3-D鱼幼虫数字化为2-D图像。此外,由此产生的数字图像保持其高分辨率 - 从而允许精确定量幼虫一侧的所有离子细胞及其平均大小和幼虫的表面积。然后可以将这些参数输入到简单的等式中,以估计相对于样本的总皮肤表面积的皮肤离子细胞离子传输能力。对于计算所有皮肤离子细胞繁琐且耗时的较大样本,我们开发了一种随机抽样方案,该方案允许在10%的皮肤表面积内计算离子细胞后进行高度准确的皮肤离子细胞估计。例如,13.7毫米长的黄鳍金枪鱼( Thunnus albacares >)样本中估计的皮肤离子细胞数为15,342,仅比整个鱼皮中计数的实际14,795个皮肤离子细胞高出3.6%( Kwan et al。>,2019)。我们相信这种方法对于未来的研究有价值,可用于量化幼虫和幼鱼以及其他小型水生生物在发育过程中的皮肤离子细胞以及应对各种环境压力(例如>,海洋变暖,缺氧和酸化)。

关键字:氯离子细胞, 富含线粒体的细胞, 皮肤, 幼虫, 渗透调节, pH调节, ATP酶, 海洋酸化


  1. PCR反应条(8 x 0.2 ml)连接平盖(Simport,目录号:T320-2N)
  2. 1.7 ml微量离心管(Corning,目录号:MCT-175-C)
  3. 15毫升塑料管(康宁牌,目录号:352196)&nbsp;
  4. 显微镜载玻片(Thermo Fisher Scientific,目录号:12-550-15)或凹面显微镜载玻片(United Scientific Supplies,目录号:CS3X11)
  5. 显微镜盖玻片(例如>,VWR,目录号:48366 045)
  6. 显微镜小培养皿托盘(MatTek Corp,目录号:P35GC-1.5-10-C)
  7. 培养皿(Corning Brand,目录号:351029)或任何其他容器
  8. 感兴趣的样本
  9. 去离子(DI)水
  10. 10x磷酸盐缓冲盐水(PBS)(Corning Incorporated,目录号:46-013-CM)
  11. 20%多聚甲醛(PFA)(Electron Microscopy Sciences,目录号:15713)
  12. 200-proof(100%)乙醇(Decon Laboratories,目录号:2701)
  13. 30%(w / v)过氧化氢(Sigma-Aldrich,目录号:H3410)
  14. α5Na + / K + ATPase(NKA)抗体(Developmental Studies Hybridoma Bank,目录号:a5-上清液)
    1. 这种单克隆抗体特异性识别来自多种水生生物的NKA,包括:骨鱼(Wilson > et al。> ,2000和2002; Yang > et al。> ,2013; Tang > et al。> ,2014; Kwan > et al。 > ,2019年),elasmobranch鱼类(Piermarini和Evans,2001; Roa > 等。> ,2014; Roa和Tresguerres,2017)和hagfish(Clifford < / em> et al。> ,2015)。>
    2. 该方案可以在感兴趣的物种中的一抗验证后适用于其他蛋白质。然而,使用不同的一抗成功定位可能需要优化此处描述的固定和免疫组化程序。>
  15. Vectastain ® Elite ® ABC HRP即用型试剂盒(Vector Laboratories,目录号:PK-7200),包括:
    1. 正常马血清阻断液
    2. 泛特异性二抗溶液&nbsp;
    3. 链霉抗生物素蛋白过氧化物酶溶液
  16. 3,3-二氨基联苯胺(DAB)过氧化物酶底物储存于4°C(Vector Laboratories,目录号:SK-4100)
  17. 叠氮化钠(Thermo Fisher Scientific,目录号:S227I-25)
  18. 1.2x磷酸盐缓冲盐水(PBS)溶液(见食谱)
  19. 4%多聚甲醛(PFA)固定液(见食谱)
  20. 3%过氧化氢溶液(见食谱)
  21. 1.0x PBS解决方案(参见食谱)
  22. DAB底物工作液(见食谱)
  23. 10%叠氮化钠储备液(见食谱)
  24. 0.02%叠氮化钠工作液(见食谱)


  1. 冰箱(4°C)和冰柜(-80°C)
  2. 化学通风橱
  3. 硼硅酸盐玻璃小瓶(Thermo Fisher Scientific,目录号:033374)
  4. 振动台(例如>,VWR,标准轨道振动器,型号1000)或旋转器(例如>,Thermo Fisher Scientific,Tube Revolver / Rotator,目录号:88881001)
  5. 转移移液管(Thermo Fisher Scientific,目录号:1368050)或适合您感兴趣标本的任何其他移液器
  6. 精细镊子(精细科学工具,目录号:26029-10)或任何不会损坏您感兴趣的样本的镊子
  7. 安装在相机上的显微镜用于较小样品的Leica DMR复合显微镜(<5 mm SL),用于较大样品的Leica MZ9.5立体显微镜(> 5 mm)
  8. 显微镜光源(Ludl Electronic Products [LEP] [12 V,100 W],目录号:99019)
  9. 显微镜安装摄像头(佳能,Rebel T3i单反相机)
  10. 兼容相机与SD卡
  11. 用于显微镜成像的抗振动台
  12. 显微镜级千分尺标准
  13. 监视显微镜摄像机的实时馈送成像
  14. 用于无线快门释放的相机遥控器(佳能RC-6)


  1. Helicon Focus(HeliconSoft,版本6.7.2或更高版本, www.heliconsoft.com
  2. Adobe Photoshop(Adobe Systems,CS3或更高版本, www.adobe.com
  3. 斐济( https://imagej.net/Fiji )或ImageJ( https://imagej.nih.gov/ij/download.html )带有单元格计数器插件( https://imagej.nih.gov/ij/plugins/cell-counter.html
  4. Adobe Illustrator(Adobe Systems,版本19.2.0或更高版本, www.adobe.com
  5. R( https://www.r-project.org




  1. 样品固定
    1. 按照批准的动物护理协议对标本进行安乐死。
    2. 将样品在固定溶液(参见配方1和2)中在旋转器/摇动台上的任何不透水容器(例如>,硼硅酸盐玻璃瓶)中孵育过夜(8-12小时),温度为4°C。
      1. 所有样品应完全浸入固定剂中。多个样品可以固定在一起。在Kwan > 等人> ,2019年,将6-10只鱼幼虫(长度为3-25毫米)固定在20毫升硼硅酸盐玻璃中的10毫升固定剂中小瓶。根据您感兴趣的样本调整容器,体积和固定剂。不需要在固定之前冲洗样品。>
      2. 如果要立即处理样品用于免疫组织化学,可以用三次10分钟PBS洗涤替换乙醇洗涤(步骤A3和A4)。>
    3. 取出固定剂,然后在摇床上将样品在50%乙醇中孵育过夜(8-12小时),温度为4°C。
    4. 去除50%乙醇,然后在70%乙醇中孵育样品,在4°C下长期储存。

  2. 整体免疫组织化学
    1. 通过用自来水转移到培养皿上1分钟开始再水化样品。
    2. 初次冲洗后,用自来水(填充至最大体积)将样品转移到塑料管(例如>,0.2 ml PCR管或1.7 ml微量离心管,具体取决于幼虫大小)中,并进一步在10分钟内孵育10分钟。旋转/摇床上的室温。
      1. 所有样品应完全浸入溶液中。在Kwan > 等人> ,2019中,将鱼幼虫(长度<20mm)在1.7ml微量离心管中温育并浸入~1ml溶液中。较大的样品将需要更大的容器和更多的解决方案(反之亦然)。相应地调整容器和溶液体积。>
      2. 在溶液更换期间,优选从容器中取出溶液 - 而不是将样品移至具有不同溶液的容器中。这最大限度地减少了对样品的损害。这适用于所有解决方案更改。>
      3. 如果使用塑料移液管更换溶液会出现问题,请考虑使用尺寸更合适的移液器(P200移液器,200 ml尖端用于0.2 ml PCR管,P1000移液器用1000μl吸头)更换溶液。根据样品和/或容器进行调整。>
    3. 除去水,然后在室温下在旋转器/摇动台上将样品在新制备的3%过氧化氢(参见配方3)中孵育10分钟。
      1. 在溶液更换期间,优选从容器中取出溶液 - 而不是将样品移至具有不同溶液的容器中。这最大限度地减少了对样品的损害。这适用于所有解决方案更改。>
      2. 如果有强烈色素沉着和/或非特异性背景染色,此步骤可延长至1小时。>
    4. 除去3%过氧化氢溶液,然后在室温下在旋转器/摇动台上将样品在正常马血清(NHS)封闭溶液中孵育15分钟。
    5. 制备一抗溶液(例如>,α5NKA抗体:在NHS封闭溶液中20ng / ml)。
    6. 去除NHS,然后在一抗溶液中孵育样品,并在室温或(优选)4°C下在旋转器/摇床上放置过夜(≥8小时)。
      注意:可以针对各种抗体验证对照修改该步骤。有关详细信息,请参阅程序K. >
    7. 去除一抗溶液,然后在室温下在旋转器/摇动台上用1x PBS(参见配方4)冲洗样品5分钟。重复两次,总共三次冲洗。
    8. 取出1x PBS洗液,然后在室温下在旋转器/摇动台上将样品在泛特异性二抗溶液中孵育30分钟。
    9. 去除泛特异性二抗溶液,然后在室温下在旋转器/摇床上用1x PBS冲洗样品5分钟。重复两次,总共三次冲洗。
    10. 取出1x PBS洗液,然后在室温下在旋转器/摇动台上将样品在链霉抗生物素蛋白过氧化物酶溶液中孵育15分钟。
    11. 去除链霉抗生物素蛋白过氧化物酶溶液,然后在室温下在旋转器/摇动台上用1x PBS冲洗样品5分钟。重复两次,总共三次冲洗。
    12. 根据制造商的说明,在单独的容器(例如>,15 ml离心管)中制备DAB底物工作溶液(参见配方5)。
    13. 取出1x PBS洗液,然后在室温下将样品在DAB底物工作溶液中孵育20秒。
      1. 该部分的敏感性高度依赖于感兴趣的组织样品和抗体。因为DAB染色是不可逆的,所以如果需要更深染色,则从20秒开始并增加孵育时间。最佳培养时间应由研究人员确定。>
      2. 孵育样本的时间较长可能会导致背景染色较深。如果这成为问题,请考虑用去离子水稀释DAB底物工作溶液。>
    14. 除去DAB底物工作溶液,然后在室温下在旋转器/摇动台上用去离子水冲洗样品10分钟。重复两次,总共三次冲洗。
    15. 储存在4°C的去离子水中直至成像。
    16. 如需长期储存(1周或更长时间),请使用PBS + 0.02%叠氮化钠(参见食谱6和7)。

  3. 光学显微镜和成像
    1. 小心地将样品转移到显微镜载玻片,凹面显微镜载玻片或玻璃底皿上,在光学显微镜下检查。
    2. 加入去离子水以防止样品变干,并用盖玻片固定样品以防止意外移动。
    3. 打开DSLR相机上的实时馈送选项。
    4. 关闭DSLR相机上的自动对焦选项。
    5. 将相机设置为手动或光圈优先模式。调整以下参数,直到图像最佳:相机ISO,相机光圈,显微镜光圈,光源亮度。
    6. 为获得最佳成像效果,请使用无线遥控器触发快门释放。
    7. 从一端(例如>,鱼标本的头部)开始,在视野内的每个焦平面(Z-堆叠)处拍摄样本。手动调整焦距,直到整个Z-stack完成(图2A-2E)。小心不要在X或Y维度上移动图像。
      1. 显微镜镜头和相机之间的图像可能不同(在实时供纸监视器上显示)。始终使用相机的焦平面并相应地拍摄图像。>
      2. 最好按连续顺序拍摄Z-stack(从前到后,反之亦然),因为它可能有助于后期处理。>
      3. 确保样品不会在图像之间的x或y平面漂移,以进行正确的后处理。>
      4. 精确的Z步(每个图像之间的距离)在样本中的范围很大。在每张图像之后,转动精细焦点,直到下一个平面对焦。在Kwan > 等> ,2019年,大多数图像每个Z-stack占用大约6-8张图像。>

      图2. Z-堆叠和聚焦堆叠的示例。在腹侧体(AE)上拍摄Z-堆叠后,通过聚焦堆叠(F)将每个图像的聚焦部分堆叠在一起。 br />
    8. 移动到下一个未拍摄的区域,注意将连续的Z-堆叠重叠至少约20%,以确保照片拼接精度(图3)。

      图3. Z-stack的重叠对于正确的照片拼接是必要的。每个Z-堆叠(A,B)应该具有~20%的重叠区域以实现最佳的照片拼接(C)。 br />
    9. 继续对Z-stack进行成像,直到拍摄完整个样本。
    10. 拍摄显微镜标准,高度约等于样品厚度的一半。

  4. 聚焦和照片拼接
    1. 可以在youtube上找到解释焦点堆叠过程的视频教程 - > 使用Helicon快速入门重点> (由HeliconSoft发布)。>
    2. 可以在youtube上找到解释照片拼接过程的视频教程 - > 如何拼接多个照片自动与Photoshop中的Photomerge一起> (由Photoshop Video Academy发布)。>
    1. 将每堆图像导入Helicon Soft并渲染焦点堆叠。
      1. 从方法C(金字塔)开始。&nbsp; >
      2. 如果图像边缘变形(光晕效果),则使用方法A(加权平均值)或方法B(深度图)进行故障排除 - 并增加“半径”值。在此处阅读有关不同方法的更多信息:> https:// www .heliconsoft.com /螺旋聚焦-主参数/ > >
      3. 可以使用其他焦点堆叠软件(> ,例如> ,Adobe Photoshop),但处理速度和结果质量可能会有所不同。>
    2. 对每堆图像重复上述步骤。保存图像。
    3. 在Adobe Photoshop中打开所有焦点堆叠图像并检查处理工件(图4)。如果存在工件,请使用“裁剪”工具(热键:C)丢弃工件。在继续之前保存所有更改。


    4. 在Adobe Photoshop中打开所有焦点堆叠图像并审查工件后,使用照片拼接工具(文件&gt;自动化&gt; Photomerge ...)。
    5. 选择“添加打开文件”(所有文件必须已保存),选择“自动”,然后选择“确定”。
    6. 检查工件或未拍摄区域。保存图片。
    7. 打开在步骤C10中取回的包含显微镜标准的图像。使用“矩形选框”工具(热键:M)测量已知宽度(例如>,1 mm)。&nbsp;
    8. 使用“画笔”工具(热键:B)填充矩形,其颜色与背景形成鲜明对比。
    9. 选择彩色矩形后,复制(Windows热键:Ctrl + C; Mac热键:Cmd + C)比例尺并粘贴(Windows热键:Ctrl + V; Mac热键:Cmd + V)到照片拼接图像上样品。保存图片。

  5. 校准用于图像分析/离子细胞定量的测量工具
    注意:可以在ImageJ中的Youtube-Set校准(由remotelab PloyU发布)中找到解释此过程的视频教程。>
    1. 在FIJI / ImageJ中打开照片拼接图像。
    2. 使用“放大镜”工具或按“+”键,放大比例尺。
      注意:在按“+”之前,将鼠标光标悬停在要放大的区域上。您也可以按“ - ”缩小。>
    3. 使用“直线”工具,在比例尺保持移位上画一条线,将线锁定在90°角。&nbsp;
    4. 在窗口顶部(Mac屏幕),单击Analyze&gt;设置比例。应该已经填充了像素的距离,因为这是您刚绘制的线条。填写“已知距离”和“长度单位”(例如>,1 mm),然后单击“确定”。
    5. 为确保准确性,请打开先前在“步骤C10”中拍摄的显微镜标准图像。使用“直线”工具,在成像的比例尺上画一条线并测量(热键:M)。如果操作正确,成像比例尺和“测量”距离之间的差异应该可以忽略不计。

  6. 表面积估计
    1. 使用FIJI / ImageJ的“直线”工具,测量(热键:M)鱼幼虫的标准长度。记录这个值。
    2. 使用“徒手选择”工具,追踪鱼幼虫以测量(热键:M)其表面积。记录这个值。
    3. 使用“徒手选择”工具,跟踪离子细胞分布可能不同的感兴趣区域(在本例中我们使用头部,身体和鳍)和测量(热键:M)。记录这些值。

  7. 计数皮肤离子细胞
    1. 该技术适用于较小的鱼样本(<5 mm标准长度)和具有非均匀分布的离子细胞的样本。>
    2. 解释此过程的视频教程可以在ImageJ的Youtube-Count项目中找到(由Timothy Spier出版)。>
    1. 打开FIJI / ImageJ。
    2. 如果使用ImageJ,请确保已安装“Cell Counter”插件(请参阅软件)。
    3. 在FIJI / ImageJ中打开照片拼接图像。
    4. 转到插件&gt;分析&gt; <细胞计数器>细胞计数器。
    5. 在“细胞计数器”窗口中,单击“初始化”,选择“类型1”,然后在感兴趣的区域中放大(“+”)。&nbsp;
    6. 使用鼠标左键单击以表示该位置的单元格。计数器将在Cell Counter窗口的“Type 1”旁边更新。记录此值。
    7. 要删除错误计数,请选中“删除模式”旁边的框。接下来,单击要删除的点。取消选中此框以继续计数。
    8. 使用“单元格计数器”窗口中的“保存标记”选项可以保存您的工作。

  8. 估计皮肤离子细胞的数量
    1. 在Adobe Illustrator中使用比例尺打开照片拼接图像。
      1. 要放大,请使用缩放工具(热键:Z)并单击。>
      2. 要缩小,请使用缩放工具(热键:Z),按住ctrl(Mac:按住Cmd)并单击。>
    2. 调整画板(热键:Shift + O)以适合照片拼接图像。如果照片拼接图像大于最大画板尺寸,则缩小图像尺寸1)单击图像,2)悬停在角落锚点上,3)按住“shift”,4)拖动以缩小图像尺寸。
    3. 将此图层的名称更改为“图像”,然后将其锁定。创建四个附加图层并按此顺序排列:鳍数,体数,头数,网格,图像(从上到下)(图5)。
      1. 您可能还有其他感兴趣的领域需要额外的图层>
      2. 图层只有在1)可见,2)解锁和3)选中时才可编辑。>
      3. 顶部的图层将覆盖底部的图层。>


    4. 切换“标尺”选项(窗口:Ctrl + R; Mac:Cmd + R)。右键单击标尺并更改为“像素”。
    5. 使用矩形工具(热键:M),在比例尺上拖动一个矩形(图6A)。
    6. 在窗口顶部朝右侧的第二行,找到并单击橙色“变换”。&nbsp;
    7. 记录“W”旁边列出的值(图6B)。此值将帮助您在像素和测量的公制单位之间进行转换。


    8. 使用在步骤F1中计算的标准长度,计算等于鱼样标准长度的2%的值。
    9. 使用矩形工具,单击(不拖动)以创建具有自定义宽度和高度尺寸的形状。将“2%标准长度”值输入“宽度”和“高度”尺寸(制作正方形)。
      注意:例如,如果标本的标准长度= 8.80 mm,则标准长度的2%为0.176 mm。>
    10. 复制并粘贴方块,直到网格覆盖整条鱼(图7)。
      1. 全选:“Windows:Ctrl + A,Mac:Cmd + A”>
      2. 复制:“Windows:Ctrl + C; Mac:Cmd + C“>
      3. 粘贴到位:“Windows:Ctrl + Shift + V; Mac:Cmd + Shift + V“>
      4. 在拖动新粘贴的单元格时,按住“Shift”键锁定X或Y平面中的方块。>
      5. 默认情况下,插图画家应“自动对齐”您的选择。如果关闭此功能,请转到偏好设置&gt;选择和锚点显示并选中“对齐点”框。>


    11. 对网格的x和y轴进行编号。
    12. 计算单个正方形的面积。记录这个值。
      注意:例如,如果标本的标准长度= 8.80 mm,则每个方格的面积为0.031 mm 2 。>
    13. 使用步骤F2和F3中的测量值,计算需要计数的平方数,以覆盖每个感兴趣区域的10%(例如>,头部,身体,鳍)。记录覆盖每个区域至少10%所需的方格数。
      1. 如果头部区域面积= 4.9 mm 2 且每个正方形为0.031 mm > 2 > ,然后:> 。向上计算并计算16个方格,以充分覆盖至少10%的头部区域。>
      2. 不同体区的分离是必要的,因为区域离子细胞密度的变化可能由于物种特异性或发育阶段特异性差异而发生。>
    14. 打开R.复制并粘贴提供的代码(参见下面的 #1 >)。在使用之前相应地调整X和Y坐标的数量。
      ### Note#1 >
      #Step-by-step instruction >
      #Step 1)在第15行中,将“59”替换为x轴上的方块数> #Step 2)在第16行中,将“17”替换为y轴上的方块数> #Step 3)突出显示所有内容(Windows:ctrl + A; Mac:cmd + A)>
      #Step 4)按住“Shift”并点击“enter”>
      #Step 5)在R Console窗口中查看输出。>

      x.coord = sample(1:59,replace = T,500)>
      y.coord = sample(1:17,replace = T,500)>
      rand.coords&lt; - data.frame(x.coord,y.coord)>
      rand.coords >


    15. 对于每个随机生成的坐标,如果方块包含以下任何一项,则取消计数:
      1. 背景空间,眩光或其他摄影文物(图8A)。
      2. 折叠,内陷,颜料,防止准确计数(图8B)。
      3. 不止一个感兴趣的区域:头部,身体,鳍(图8C)。


    16. 如果方格符合上述标准,则标记为计数。
      1. 要快速指定计数方块,请使用“铅笔工具”(热键:N)来刻划标记。>
      2. 为避免意外移动网格或底层图像,请务必同时锁定网格和图像层(参见图5B)。>
    17. 计算Adobe Illustrator中的单元格。对于每个区域,记录随机选择的网格方格中存在的离子细胞数量。
    18. 如果网格线碰巧与免疫染色细胞重叠,则仅计数与正方形的顶部和右侧交叉的细胞。如果两个方格是彼此相邻的随机选择,这将确保不对细胞进行重复计算。
      注意:如果网格线太粗或单元格直接位于线条下方,请尝试选择网格图层(点击网格图层中最右边的圆圈;图5D)并将“笔触”值减小为“0.5” “或者将不透明度降低到50%以便更好地观看。>
    19. 继续,直到每个区域(头部,身体,鳍)的至少10%被计算(图9)。

      图9.黄鳍金枪鱼标本和早期幼体皮肤离子细胞的定量方法(标准长度> 5 mm)。通过强烈的Na + / K鉴定离子细胞 + -ATPase免疫染色,并在头部(绿色),体(粉红色)和鳍(蓝色)区域内的重叠网格的随机取样框内计数。虚线白线勾勒出由于严重色素沉着而未采样的区域,从而阻止了精确的离子细胞计数。估计的离子细胞计数(15,342细胞)用手工离子细胞计数(无色区域; 14,795个细胞)进行研磨。该幼虫的估计相对离子细胞面积为3.75%。图来自Kwan 等人>,2019。

    20. 使用以下等式计算估计值:

      1. 区域头部,身体或鳍 =头部,身体或鳍区域的测量区域(步骤F3)。>
      2. 离子细胞头部,身体或鳍 =每个区域计算的离子细胞总和(步骤H17)。>
      3. Square head,body或fins =每个区域计算的平方数(步骤H13)。>
      4. Area square =每个方块的大小=(标准长度* 0.02) 2 (步骤H12)。>

  9. 量化离子细胞大小
    1. 打开ImageJ并设置比例尺,如过程E中所述。
    2. 使用“徒手选择”工具,放大(热键:+)并跟踪单个皮肤离子细胞以估计它们的相对表面积。
    3. 重复步骤I2,直到您测量了来自不同感兴趣区域的20个细胞(例如>,头部,身体和鳍)。
    4. 使用以下等式计算平均离子细胞大小:

      区域 head,body或fins 在步骤F3中计算。
      在步骤F2中计算面积 total 。

  10. 量化相对离子细胞面积


  11. 方法控制和抗体验证
    1. 非特异性二级和三级抗体结合对照:通过在步骤B6期间省略一级抗体(在4℃下在NHS封闭溶液中过夜)将几个鱼标本指定为阴性对照。与原代NKA抗体孵育的鱼标本应导致阳性染色(图10A)。相反,没有一抗的鱼样本应该导致没有染色(图10B)。
      1. 执行此阴性对照步骤是识别和避免目标生物体上存在的天然色素的好方法(步骤H15)。>
      2. 如果需要,可以重新培养和利用阴性对照鱼样品。只需重复所有程序B,重新免疫所述样品。>

      图10.免疫组织化学的阴性对照与(A)和没有(B)Na + / K + -ATPase(NKA)抗体。在顶部幼虫中看到的棕色沉淀物表示NKA存在于皮肤表面上,而在底部幼虫上没有褐色染色,成功地产生阴性对照。注意在估算过程中应避免在两个试样中含有天然色素的区域。

    2. 通过蛋白质印迹的一抗抗体特异性:快速冷冻组织(-80°C)应与固定组织一起进行蛋白质印迹取样,这样可以产生一条接近目标蛋白质预测大小的单一条带(图11)。有关协议和其他详细信息,请参阅Kwan et al。>(2019)中的补充材料2。

      图11. Na + / K + <的Western印迹在黄鳍金枪鱼组织中用抗-NKA单克隆抗体进行Western印迹,得到单个~108 kDa的条带,与预测的蛋白质大小相匹配。 -ATPase(NKA)在黄鳍金枪鱼组织中的表达。图改编自Kwan 等人>,2019。


离散细胞计数(步骤H20),平均离子细胞大小(步骤I4)和相对离子细胞面积(步骤J1)数据最好在散点图中描述并使用回归模型进行解释。在Kwan et al。>,2019年,我们使用线性回归检验了标准长度和孵化后天数的相对离子细胞面积的关系。线性回归的斜率解释了线的陡度,零假设为零,而R 2 描述了回归的拟合优度。可能适用的其他发展指标包括累积的热单位和样本的总长度。发展可能并不总是线性的;因此,研究人员应根据您的研究领域选择自己的回归。如果实验包括多种处理或物种,则应使用 t > - 测试(总共2次处理)或协方差分析(3次或更多次处理)来检查回归斜率的比较。有关其他详细信息,请参阅Zar(2014)。


  1. 1.2x磷酸盐缓冲盐水(PBS)溶液(1L)
  2. 4%多聚甲醛(PFA)固定液(10 ml)
    湾将2ml PFA储备溶液加入8ml 1.2x PBS溶液(配方1)中,在1M PBS中终浓度为4%PFA。反转混合
    d。 4%PFA固定液可在4°C下储存长达一周
  3. 3%过氧化氢溶液(10毫升)
  4. 1.0x PBS溶液(1升)
  5. DAB底物工作液(与制造商说明相同)(~5 ml)
  6. 10%叠氮化钠原液(10毫升)
  7. 0.02%叠氮化钠工作液(10毫升)


该方案对应于Kwan 等人>(2019)中提出的方法和实验。用于创建和优化该方案的幼虫黄鳍金枪鱼和白鲈鱼样品( Atractoscion nobilis >)由美洲热带金枪鱼委员会(IATTC)和Hubbs海洋世界研究所(HSWRI)提供。我们感谢Jeanne Wexler(IATTC)对原始项目的贡献,以及对黄鳍金枪鱼幼虫的饲养和取样。我们感谢Diomedes Ballesteros,Lina Castillo,Susana Cusatti,Danisin Dominguez,Agustin Ortega,Daniel Perez,Dario Ramires,Daniel Solis,Luis Tejada和Carlos Vergara协助在巴拿马Achotines实验室生产和饲养黄鳍金枪鱼幼虫。我们感谢Daniel Margulies(IATTC)和Vernon Scholey(IATTC)开发黄鳍金枪鱼饲养方法,并监督Achotines实验室的产卵和幼虫饲养。我们感谢Erica Brombay-Fanning(HSWRI),Eric McIntire(HSWRI),Sabrina Sobel(HSWRI)和Christy Varga(HSWRI)对幼虫白鲈鱼的白鲈幼虫进行饲养和取样。我们感谢Mark Drawbridge(HSWRI)监督美国卡尔斯巴德Leon R. Raymond海洋鱼类孵化场的产卵和幼体饲养。作者要感谢Greg Rouse博士使用显微镜和相机设备,Sabine Faulhaber在扫描电子显微镜方面提供技术帮助,感谢Taylor Smith在解剖和成像方面的帮助。 G.T.K.得到了圣地亚哥奖学金和国家科学基金会研究生研究奖学金计划的支持。




所有实验均经Scripps海洋研究所/加利福尼亚大学圣地亚哥动物护理委员会根据协议no。 S10320符合机构动物护理和使用委员会(IACUC)关于实验动物护理和使用的指南。


  1. Clifford,A.M.,Goss,G.G.,Roa,J。N.和Tresguerres,M。(2015)。 hagfish中的酸/碱和离子调节。在: Hagfish Biology >。科学出版社,博卡拉顿,第277-292页。
  2. Evans,D.H.,Piermarini,P.M。和Choe,K.P。(2005)。 多功能鱼鳃:气体交换,渗透调节,酸碱调节和排泄的主要部位含氮废物。 Physiol Rev > 85(1):97-177。
  3. Fridman,S.,Bron,J.E。和Rana,K.J。(2011)。 尼罗罗非鱼Oreochromis niloticus早期生活阶段氯离子细胞位置和形态的发育变化适应于淡水和微咸水。 J Fish Biol > 79(3):597-614。
  4. Hiroi,J.,Kaneko,T.,Seikai,T。和Tanaka,M。(1998)。 日本牙鲆( Paralichthys olivaceus < / em>)幼虫。 Zoolog Sci > 15(4):455-460。
  5. Hiroi,J.,Kaneko,T。和Tanaka,M。(1999)。 卵黄囊膜中氯离子细胞形态的体内>顺序变化莫桑比克罗非鱼( Oreochromis mossambicus >)在海水适应期间的胚胎和幼虫。 J Exp Biol > 202 Pt 24:3485-3495。
  6. Hwang,P。P.(1989)。 硬骨鱼幼虫中氯离子细胞的分布。 J Morphol > 200(1):1-8。
  7. Katoh,F.,Shimizu,A.,Uchida,K。和Kaneko,T。(2000)。 海水适应性Killifish, Fundulus heteroclitus <早期生命阶段氯离子细胞分布的变化< / em>。 Zoolog Sci > 17(1):11-18。
  8. Kwan,G.T.,Wexler,J.B.,Wegner,N.C。和Tresguerres,M。(2019)。 黄鳍金枪鱼的皮肤和鳃离子细胞和形态的发生变化( Thunnus albacares >)幼虫。 J Comp Physiol B > 189(1):81-95。
  9. Piermarini,P。M.和Evans,D。H.(2001)。 euryhaline黄貂鱼鳃中液泡质子-ATP酶B亚基的免疫化学分析( Dasyatis sabina >):盐度的影响以及与Na + / K + -ATPase的关系。 J Exp Biol > 204(Pt 19):3251-3259。
  10. Roa,J。N.,Munevar,C。L.和Tresguerres,M。(2014)。 饲喂诱导液泡质子ATP酶和pendrin易位到豹鲨膜上( Triakis semifasciata >)富含线粒体的鳃细胞。 Comp Biochem Physiol A Mol Integr Physiol > 174:29-37。
  11. Roa,J。N.和Tresguerres,M。(2017)。 碳酸氢盐敏感的可溶性腺苷酸环化酶存在于多个鲨鱼组织的细胞质和细胞核中。 Physiol Rep > 5(2)。
  12. Tang,C.H.,Leu,M.Y.,Yang,W.K。和Tsai,S.C。(2014)。 探讨蛋白质质量控制和绿色鳃鳃的渗透调节机制 >以减少盐度。 Fish Physiol Biochem > 40(5):1533-1546。
  13. Varsamos,S.,Diaz,J.P.,Charmantier,G.,Blasco,C.,Connes,R。和Flik,G。(2002)。 欧洲鲈鱼胚胎发育过程中氯细胞的位置和形态, Dicentrarchus labrax >。 Anat Embryol(Berl)> 205(3):203-213。
  14. Varsamos,S.,Nebel,C。和Charmantier,G。(2005)。 胚胎鱼体内渗透调节的开发:综述。 Comp Biochem Physiol A Mol Integr Physiol > 141(4):401-429。
  15. Wilson,J.M.,Randall,D.J.,Donowitz,M.,Vogl,A.W。和Ip,A.K。(2000)。 离子转运蛋白免疫定位于mudskipper中的鳃上皮线粒体富含细胞( Periophthalmodon schlosser > i >)。 J Exp Biol > 203(Pt 15):2297-2310。
  16. Wilson,J.M.,Whiteley,N.M。和Randall,D.J。(2002)。 海水驯化过程中银大麻鳃上皮细胞的离子调节变化。 Physiol Biochem Zool > 75(3):237-249。
  17. Yang,W.K.,Kang,C.K.,Chang,C.H。,Hsu,A.D.,Lee,T.H。和Hwang,P.P。(2013)。 在咸水青蛙 Oryzias dancena >中的鳃FXYD蛋白的表达谱:a 用于渗透率调节研究的潜在咸水鱼模型。 PLoS One > 8(1):e55470。
  18. 扎尔,J.H。(2014)。 比较简单线性回归。在:生物统计分析>。 在:Pearson,H。(编辑)。 pp:387-404。
  • English
  • 中文翻译
免责声明 × 为了向广大用户提供经翻译的内容,www.bio-protocol.org 采用人工翻译与计算机翻译结合的技术翻译了本文章。基于计算机的翻译质量再高,也不及 100% 的人工翻译的质量。为此,我们始终建议用户参考原始英文版本。 Bio-protocol., LLC对翻译版本的准确性不承担任何责任。
Copyright: © 2019 The Authors; exclusive licensee Bio-protocol LLC.
引用:Kwan, G. T., Finnerty, S. H., Wegner, N. C. and Tresguerres, M. (2019). Quantification of Cutaneous Ionocytes in Small Aquatic Organisms. Bio-protocol 9(9): e3227. DOI: 10.21769/BioProtoc.3227.