Login processing...

Trial ends in Request Full Access Tell Your Colleague About Jove

Medicine

A High-Throughput In Situ Method for Estimation of Hepatocyte Nuclear Ploidy in Mice

doi: 10.3791/60095 Published: April 19, 2020

Summary

We present a robust, cost-effective, and flexible method for measuring changes in hepatocyte number and nuclear ploidy within fixed/cryopreserved tissue samples that does not require flow cytometry. Our approach provides a powerful sample-wide signature of liver cytology ideal for tracking the progression of liver injury and disease.

Abstract

When the liver is injured, hepatocyte numbers decrease, while cell size, nuclear size and ploidy increase. The expansion of non-parenchymal cells such as cholangiocytes, myofibroblasts, progenitors and inflammatory cells also indicate chronic liver damage, tissue remodeling and disease progression. In this protocol, we describe a simple high-throughput approach for calculating changes in the cellular composition of the liver that are associated with injury, chronic disease and cancer. We show how information extracted from two-dimensional (2D) tissue sections can be used to quantify and calibrate hepatocyte nuclear ploidy within a sample and enable the user to locate specific ploidy subsets within the liver in situ. Our method requires access to fixed/frozen liver material, basic immunocytochemistry reagents and any standard high-content imaging platform. It serves as a powerful alternative to standard flow cytometry techniques, which require disruption of freshly collected tissue, loss of spatial information and potential disaggregation bias.

Introduction

or Start trial to access full content. Learn more about your institution’s access to JoVE content here

Hepatocytes in the mammalian liver can undergo stalled cytokinesis to produce binuclear cells, and DNA endoreplication to produce polyploid nuclei containing up to 16N DNA content. Overall cellular and nuclear ploidy increase during postnatal development, ageing and in response to diverse cellular stresses1. The process of polyploidization is dynamic and reversible2, although its precise biological function remains unclear3. Increased ploidy is associated with reduced proliferative capacity4, genetic diversity2, adaptation to chronic injury5 and cancer protection6. Hepatocyte ploidy alterations occur as a result of altered circadian rhythm7, and weaning8. Most notably, the ploidy profile of the liver is altered by injury and disease9, and compelling evidence suggests that specific ploidy changes, such as increased ≥8N nuclei or loss of 2N hepatocytes, provide useful signatures for tracking non-alcoholic fatty liver disease (NAFLD) progression3,10, or the differential impact of viral infections11.

In general terms, liver injury and regeneration are associated with increased hepatocyte cell size and nuclear area12, together with reduced overall numbers of hepatocytes, particularly those with 2N DNA content10,11. Parenchymal injury in the liver is also frequently accompanied by expansion of non-parenchymal cells (NPCs), including stromal myofibroblasts, inflammatory cells and bipotent liver progenitor cells. High-throughput methods that provide a quantitative cytological profile of parenchymal cell number and nuclear ploidy, whilst also accounting for changes in NPCs, therefore have considerable potential as research and clinical tools to track the response of the liver during injury and disease. Compelling recent in situ analysis of ploidy spectra in human samples of hepatocellular carcinoma also demonstrate that nuclear ploidy is dramatically increased within tumors and is specifically amplified in more aggressive tumor subtypes with reduced differentiation and loss of TP5313. Hence, there is a strong possibility that methodological advances in quantitative assessment of nuclear ploidy will assist in future prognostic profiling of liver cancer.

In this protocol, a flexible high-throughput methodology for the comparative analysis of mouse liver tissue sections is described, which provides detailed cytometric profiling of hepatocyte numbers, the NPC response and an internally calibrated method for estimating nuclear ploidy (Figure 1). Hepatocytes are distinguished from NPCs by hepatocyte nuclear factor 4 alpha (HNF4α) immunolabelling, prior to characterization of nuclear size and nuclear morphometry. "Minimal DNA content" is estimated for all circular nuclear masks by integrating mean Hoechst 33342 intensity (a proxy for DNA density) with interpolated three-dimensional (3D) nuclear volume. Hepatocyte minimal DNA content is then calibrated using NPCs to generate a nuclear ploidy profile.

Image acquisition, nuclear segmentation and image analysis are performed using high-content imaging, enabling large areas of two-dimensional (2D) liver sections containing tens of thousands of cells to be screened. A custom-written program is provided for automated post-processing of high-content image analysis data to produce a sample-wide ploidy profile for all circular hepatocyte nuclei. This is performed using free to download software to calculate nuclear ploidy based on stereological image analysis (SIA)10,11,14,15. The SIA methodology has been previously validated by flow cytometry as an accurate, albeit laborious, method for estimating hepatocyte nuclear ploidy in the liver14, assuming circular nuclear morphology and a monotonic relationship between nuclear size and DNA content. In this protocol, both nuclear parameters are measured by assessment of nuclear morphometry and Hoechst 33342 labelling. Calculation of "minimal DNA content" for each nuclear mask is followed by calibration of hepatocyte nuclear ploidy using NPCs, which have a known 2−4N DNA content and therefore serve as a useful internal control.

Compared to conventional flow cytometry methods16 the approach described enables hepatocyte nuclear ploidy to be assessed in situ and does not require access to fresh tissue or disaggregation methods that can bias outcomes and be difficult to standardize. As with all SIA-based approaches, nuclear ploidy subclasses >2N are underrepresented by 2D sampling due to the sectioning of larger nuclei outside of the equatorial plane. The tissue-wide ploidy profile also describes minimum DNA content for all circular hepatocyte nuclear masks, and does not directly discriminate between mononuclear hepatocytes and binuclear cells that have two discrete ("non-touching") nuclei of the same ploidy. However, the simplicity of this protocol allows considerable scope for it to be adapted to account for additional parameters such as internuclear spacing or cell perimeter analysis, that would facilitate identification of binuclear cells providing a more detailed assessment of cellular ploidy.

Subscription Required. Please recommend JoVE to your librarian.

Protocol

or Start trial to access full content. Learn more about your institution’s access to JoVE content here

All animal experiments were previously approved by the CIPF ethics committee. Mice were housed in a pathogen-free facility at the Centro de Investigación Príncipe Felipe (Valencia, Spain), registered as an experimental animal breeder, user, and supply centre (reg. no. ES 46 250 0001 002) under current applicable European and Spanish animal welfare regulations (RD 53/2013).

1. Tissue harvesting and sample preparation

NOTE: This protocol describes how to freeze tissue without prior fixation or cryopreservation. For previously fixed/cryopreserved samples proceed to section 2 and omit step 3.1. All analyses have been performed using adult female C57BL/6 mice aged 12−16 weeks.

  1. Sacrifice animals by fentanyl/pentobarbital intraperitoneal injection followed by cervical dislocation. With the mouse facing ventral side up, open the abdominal cavity and expose the liver by grasping the skin with tweezers and performing a vertical incision from the base of the lower abdomen to the base of the sternum using surgical scissors.
  2. Carefully remove the gallbladder using fine tweezers, dissect out the liver and rinse the selected liver lobule in a 10 cm Petri dish plate filled with phosphate-buffered saline (PBS).
    NOTE: It is recommended to compare the same liver lobe for each animal, in this case the median lobe was used.
  3. Fill a labelled cryomold with optimal cutting temperature (OCT) medium at room temperature (RT). Avoid OCT bubbles. If they appear, push them to the edge of the mold using a needle or pipette tip.
  4. Embed liver lobule into a filled OCT cryomold and immediately place it on dry ice to ensure rapid freezing. Store cryomolds at -80 °C until cryosectioning.

2. Cryosectioning

  1. Transport cryomolds on dry ice to avoid tissue degradation. Prior to cryosectioning equilibrate inside cryostat set to -20 °C for 20 min.
  2. Eject sample by applying pressure to the base of the plastic cryomold. Apply liquid OCT to warm sample disk at RT, position in cryostat and attach an OCT embedded liver sample. Apply gentle pressure and wait 3 min for OCT to freeze ensuring the sample sticks to the disk.
    NOTE: Avoid handling the sample with fingers as much as possible to evade tissue degradation.
  3. Lock the sample into the arm of the cryostat and adjust the orientation so that the edge of the sample is parallel with the cryostat blade. Cut into the sample until tissue is reached.
  4. Section the sample at 6 µm thickness. Place a labelled polyamide-coated slide over the sample for 5 s to let the sample stick onto the slide. Place the slide at RT for 3−5 min, then, for best results, proceed directly to section 3.
    NOTE: For processing of multiple fresh-frozen samples reproducible results have been obtained by temporarily storing slides in a slide box on dry ice until all samples have been processed. When using this approach allow all slides to equilibrate to RT before proceeding to section 3. Formalin fixed paraffin embedded (FFPE) samples can be used, although background autofluorescence is increased by this method. To proceed from FFPE samples, section at 4 µm. Mount by catching sections from 40 °C water bath on polyamide-treated slides. Heat slides for 1 h at 60 °C, then deparaffinize by serial RT washes (5 min) in Coplin jars containing xylene (x2), ethanol 100% (x2), 96% (x2), 70% (x1) and dH2O (x1). To expose antigens place slides in citrate buffer for 20 min at 90 °C before tempering slides in PBS at RT. Proceed to step 3.2.

3. Fluorescence immunolabelling

  1. Fix tissue sections in a fume hood by applying 1 mL of 4% paraformaldehyde (PFA) in PBS for 10 min at RT. Transfer slides to a PBS filled Coplin jar and wash for 3 min using gentle agitation (repeat 3x).
    NOTE: From now until the end of the immunostaining process, avoid drying of the sample.
  2. Dry the area around each tissue section and surround using a hydrophobic pen. Permeabilize with 0.5% nonionic surfactant (i.e., Triton X-100) in PBS for 15 min at RT. Then wash in PBS filled Coplin jar for 3 min using gentle agitation (repeat 2x).
  3. Block using a filtered solution of 1% bovine serum albumin (BSA), 5% horse serum, 0.2% nonionic surfactant in PBS (for at least 1 h at RT).
  4. Incubate with primary HNF4α antibody diluted in blocking buffer over night at 4 °C in a dark humid staining chamber (see Table of Materials for antibodies and specific dilutions).
  5. Place slides into a PBS filled Coplin jar and wash for 3 min using gentle agitation (repeat 4x).
  6. Incubate with Alexa-488 conjugated secondary antibody and Hoechst diluted in filtered 1% BSA and 0.2% nonionic surfactant in PBS for 2 h at RT in a dark humid staining chamber (see Table of Materials for antibodies and specific dilutions).
  7. Place slides into a PBS filled Coplin jar and wash for 3 min using gentle agitation (repeat 4x). Wash in ddH2O for 3 min using gentle agitation (repeat 2x).
  8. Mount slides by placing two drops of fluorescent mounting media on a coverslip (24 x 60 mm) and laying slides over it, eliminating bubbles by applying gentle pressure. For long-term storage, seal coverslip at edges with clear nail polish and store in the dark at 4 °C.
  9. Before proceeding, check slides using a conventional fluorescence microscope to ensure good fixation and immunolabeling.
    NOTE: See Figure 2A,B for expected results.

4. Fluorescence image acquisition

NOTE: For this step, a high-content imaging platform (Table of Materials) is required that supports automatic fluorescence image acquisition.

  1. Turn on the imaging system and open a new acquisition protocol.
  2. Select the 10x objective, note the area of the field of view (in this case 0.6 mm2).
  3. Set parameters to acquire fluorescence images using the appropriate excitation and emission filters (as per step 3.6). For Hoechst and Alexa-488, select "DAPI" and "GFP" channels with 390/18 and 438/24 nm excitation and 432.5/48 and 475/24 nm emission, respectively.
  4. Focus the sample and ensure signal intensity is non-saturating. Ensure that image capturing is done with the same exposure time for all the images or use a system where the intensity of fluorescence is corrected for the exposure time.
  5. Scan sample and acquire sufficient images to obtain complete coverage of the tissue section (approximately 20−50 fields of view, depending on the sample size).
  6. Review the image database, manually eliminating (i) poorly focused fields, (ii) those at the borders of each tissue section (to avoid biasing cell density calculations), and (iii) those containing folded/physically damaged areas of the tissue section if present.

5. Automated fluorescence image analysis

NOTE: This step requires appropriate image analysis software (Table of Materials) capable of: (1) automatically identifying Hoechst labelled nuclei within images at 405 nm (nuclear segmentation), (2) assessing mean Hoechst nuclear intensity and morphometry, and (3) threshold analysis to determine the +/- status of nuclear fluorescence at 488 nm (HNF4α). Some basic operator training/expertise is required to visually assess and adjust segmentation and thresholding parameters within the program to ensure that nuclei and HNF4α+/- status are optimally gated (Figure 2).

  1. In the image analysis software, open the acquisition file containing Hoechst (405 nm) and HNF4α (488 nm) images from step 4.5, and create a new analysis protocol.
  2. Define wavelengths to be used for nuclear segmentation (Hoechst, 405 nm) and for hepatocyte/NPC threshold analysis (HNF4α, 488 nm).
  3. Adjust the software's nuclear segmentation parameters (such as "minimum nuclear area" and nuclear detection "sensitivity") to ensure nuclei are optimally segregated.
    NOTE: Good segmentation of hepatocytes should be prioritized over that of NPCs. Hepatocyte nuclei are characteristically rounded (interquartile size range: 40−64 µm2). NPC nuclei, such as those of sinusoidal endothelia, are flattened/elliptical or irregular in shape and generally smaller and more closely packed than those of hepatocytes (interquartile size range: 30−43 µm2). For mouse liver, minimal nuclear area ≥23 µm2 and detection "sensitivity" of 65% were used (see Figure 2C,D for expected results). Sensitivity determines how pixel clusters are recognized as individual nuclei based on their intensity and should be empirically tested for each sample set by the user before proceeding with automated image analysis.
  4. Modify the threshold intensity at 488 nm to ensure optimal gating of hepatocytes (HNF4α+) and non-parenchymal cells (HNF4α-).
    NOTE: See Figure 2C,D for expected results. The value of threshold intensity is relative and will depend on staining efficiency and acquisition settings such as laser intensity. It should therefore be standardized by the user. Use known HNF4α- cells such as endothelial cells and periportal NPCs as an internal negative control and binuclear hepatocyte nuclei as a positive reference for staining. Test the analysis parameters using a small number of images to ensure good nuclear segmentation and intensity threshold segregation before applying analysis parameters to the entire dataset.
  5. Select the following nuclear parameters to be quantified: (1) nuclear area based on Hoechst staining (µm2), (2) mean nuclear Hoechst intensity (RU), (3) nuclear elongation factor (mean ratio of the short axis of the nucleus to the long axis of the nucleus, where a centre-symmetric [non-elongated] object has a value of 1, (4) Nuc 1/(form factor), mean nuclear "roundness" index calculated by perimeter 2/(4π x area). Values range from 1 to infinity, where 1 is a perfect circle, (5) HNF4α status (positive-1 or negative-0), and (6) nuclear x/y coordinates based on "center of gravity" (cg), a method for locating the center of the object from greyscale images with sub-pixel precision.
  6. Run the analysis for all sample datasets and export numerical data from step 5.5 to spreadsheet software.

6. Data analysis

NOTE: The data analysis step can be performed using any standard spreadsheet software.

  1. Calculate hepatocyte and non-hepatocyte cell numbers.
    1. Calculate the total area of liver section analyzed for each sample by multiplying the number of fields of view by the area of the field of view (step 4.2).
    2. Working with spreadsheet files generated for each liver section, filter the data by selecting only HNF4α+ nuclei. Calculate the total number of HNF4α+ nuclei analyzed and divide this by the total area analyzed to obtain mean hepatocyte density for each sample (Figure 2F).
    3. Perform the same calculation for non-parenchymal cells by filtering the spreadsheet for HNF4α- cells (Figure 2E).
  2. Calculate hepatocyte nuclear size distribution.
    1. Using spreadsheet software, filter data to select only HNF4α+ nuclei.
    2. Plot values of nuclear area in a histogram (Figure 2G). Set the bin width to 5 µm2.
      NOTE: Frequency values can be corrected for area (nuclei/mm2) as per step 6.1.1.
  3. Perform hepatocyte nuclear ploidy analysis.
    NOTE: The spreadsheet data from step 5.6 are used to generate a nuclear ploidy profile for each sample. This process has been automated and can be performed using a custom written software that is freely available to download with supporting information and demonstration datasets at https://github.com/lukeynoon (see Supplemental Files). Source code is provided for users who wish to adapt the methodology. A description of the algorithm, together with instructions for installation and use are outlined below. The program uses spreadsheet data to automatically separate hepatocyte nuclei into two groups; (1) those with "simple" circular nuclei and (2) "complex" non-circular nuclei representative of binuclear cells with >2c ploidy. The minimal nuclear DNA content (a function of nuclear area and DNA density) is next calculated for all "simple" nuclei. A subsequent step then automatically calibrates HNF4α+ hepatocyte nuclear ploidy using HNF4α- nuclei as a known 2−4N internal control.
    1. Download and install software.
      1. Download the packaged application from: https://github.com/lukeynoon
      2. Launch MATLAB. Navigate to the APP tab of the toolstrip, click Install App and open the downloaded application termed "Ploidy_Application.mlappinstall". A message will appear to confirm the successful installation.
        NOTE: The application is now ready for use and will remain in the APP tab of the toolstrip.
    2. Format input data.
      NOTE: Prior to automated nuclear ploidy analysis, all spreadsheet files containing high content imaging data (step 5.6) should be stored and formatted according to the following instructions.
      1. In each exported data file (.XLS 97-2004 workbook) from step 5.6, include a sheet termed "Cell measures" containing all the data required for the ploidy analysis set out in columns (Figure 3A). Ensure that the spreadsheet layout including column header names remains unchanged from that of Figure 3A, because the analysis method finds the correct column data by searching for these names (see demonstration datasets in Supplemental Files for reference). If for example, high-content image analysis software does not produce a "Light flux" column (Figure 3A), manually insert a "Light flux" column in the same location, i.e., column K and fill it with zeros.
      2. For each experimental condition (e.g., "Injured-d14"), provide a control dataset, which will be used to calculate the internal control for 2−4N nuclear ploidy calibration (step 6.3.4.3). Here, select liver samples from untreated adult littermates ("Control-d0"; Figure 3B−D).
      3. For biological replicates (per condition), store each spreadsheet in its own folder (as in Figure 3B). Name the folder prefixes incrementally, e.g., "Sample1, Sample2, Sample3… SampleN", as per the filenames contained within. Hence, every dataset folder (e.g., "Control-d0") should contain a series of subfolders ("Sample1", "Sample2", etc.) each containing a spreadsheet file of the same corresponding name.
    3. Run the application.
      1. Within MATLAB, launch the "Ploidy_Application" by clicking on the icon within the MY APPS tab of the toolstrip (Figure 3C). The Ploidy_Application graphical user interface (GUI) will appear (Figure 3C).
      2. Click the Path to control data button to navigate to the folder in which the control data replicates reside (e.g., "Control-d0"). This data path will then appear in the interface (e.g., /Users/Desktop/Control-d0).
      3. Next, in "folder prefix" type the name to be given to the output files (e.g., "Sample").
        NOTE: This prefix can be changed to any text, provided that the folders and filenames remain incrementally named.
      4. Click the Path to other data button and navigate to the folder in which the comparative data replicates reside (e.g., "Injured-d14"). This data path will then appear in the interface (e.g., /Users/Desktop/Injured-d14).
      5. Click Run!. When the analysis is complete, the status bar will read "Analysis Complete!..".
        NOTE: The application will report, for each sample, stratification of "simple" nuclei into ≤2n, 2n−4n, 4n−8n and 8n+ in terms of absolute counts and as a percentage of total (Figure 3D). These files will be automatically saved in each sample folder as: "Count_2n.txt", "Count_2n_to_4n.txt", "Count_4n_to_8n.txt", "Count_8n_and_higher.txt", "Percentage_2n.txt", "Percentage_2nto4n.txt", "Percentage_4nto8n.txt", "Percentage_8n_and_higher.txt". The Ploidy_Application will automatically save a list for each sample, of all the individual ploidy estimates for "simple" hepatocyte and non-hepatocyte nuclei in "Ploidy_All_Hepatocytes.txt" and "Ploidy_NonHepatocytes.txt". For the control dataset, the method also saves the minimal DNA content thresholds calculated for stratification of ploidy (see step 6.3.4.3.7) in a file named "Normalised_Thresholds_Control". Finally, the application will produce a folder for both the control and the selected comparative condition data termed "Summary". This folder contains two subfolders, "Ploidy" and "Stratification" which contain the averages of all samples provided (Figure 3D).
    4. Description of the methodology.
      NOTE: The following section describes in detail the methodology used by the Nuclear Ploidy Analysis software. If the user chooses not to use the application, these steps can be followed using spreadsheet software to calculate the nuclear ploidy profile manually.
      1. Separate nuclei into "simple" or "complex" according to nuclear morphometry.
        1. Calculate a "circularity index" for all nuclei, defined as the nuclear "elongation factor" divided by the "Nuc 1/(form factor)", where a value of 1.0 indicates a perfect circle.
          NOTE: "Nuclear elongation" and "Nuc 1/(form factor)" are two discrete measures of an object's "circularity" that assess complementary, non-overlapping morphometric criteria. The former measures the long- and short-axes of an object, while the latter compares the length of perimeter of an object to that of its area. To strengthen the definition of nuclear circularity used in this protocol, these two measurements have been combined into a single "circularity index". A previous approach to estimate nuclear ploidy using the described methodology used only nuclear elongation17. While acceptable results were obtained using this approach, the authors have observed that a composite "circularity index" improves discrimination of manually selected nuclei from mononuclear and binuclear hepatocytes (data not shown).
        2. Classify nuclei with a circularity index ≤ 0.8 as "complex" and those > 0.8 as "simple".
      2. Estimate "minimal" DNA content (m) for all "simple" nuclei.
        1. Calculate the nuclear radius (r) using the formula:
        2. Calculate nuclear volume (v) using the volume of a sphere formula:
        3. Generate a relative value for minimal DNA content (m) using the formula:
      3. Calibrate the dataset using the NPC (HNF4α-) nuclei as an internal 2−4N control.
        NOTE: NPCs have a 2−4N DNA content depending on cell cycle status. Hence, the mean value of NPC "minimal" DNA content (NPCm) increases with injury (Figure 4A). Calibration error is minimized by establishing an upper limit of NPCm representing a 4c threshold (Figure 4B).
        1. Within the spreadsheet, select only NPC nuclei with values for "m" that lie within 1 standard deviation (SD) of the mode (this filters out noise from possible segmentation error).
        2. Within this filtered range, examine nuclear areas and their corresponding mean Hoechst intensities (Figure 4C).
        3. Estimate the smallest nuclear area within this filtered range with maximal nuclear Hoechst intensity (i.e., the point at which the line of the curve changes direction in the filtered dataset as illustrated by the red circle in Figure 4C). This value represents a 2N−4N transitional state (t) above which sampling of 4c nuclei predominates over 2c nuclei, resulting in a maxima of mean Hoechst intensity.
          NOTE: This value is automatically determined by the software; however, spreadsheet users can manually select this point as the transitional size.
        4. Calculate the minimal DNA content represented by this transitional size (tm) by following step 6.3.4.2.
        5. To estimate the 4N shoulder of the NPCm dataset, add 1 SD to the value of tm. The resulting number (Figure 4B) describes the upper limit of NPC minimal DNA content to be used for nuclear ploidy stratification (S4c).
        6. Repeat steps 6.3.4.3.1−6.3.4.3.5 for all "control" samples.
          NOTE: For example, in Figure 3, uninjured control livers ("Control-d0") are used as a control condition.
        7. Calculate an average 4c stratification threshold (S4c) for "control" samples and use this to extrapolate the 2c (S2c) and 8c (S8c) boundaries for minimal DNA content (m). Stratification thresholds are automatically generated and stored by the software (step 6.3.3.3).
          NOTE: Depending on the study design, the average stratification threshold values may be calculated for each condition or for specific conditions (e.g., healthy control liver). However, the Nuclear Ploidy Analysis software requires that one of a set of 2 files is designated as "control" for the purposes of calculating relative ploidy values.
        8. Calculate a ploidy value for all nuclei using the S2c value generated in step 6.3.4.3.7 as per:
        9. Stratify "simple" hepatocyte (HNF4α+) nuclei into 2c/4c/8c/>8c brackets according to the following criteria: "2c" HNF4α+ = p ≤ 2; "4c" HNF4α+ = 2 < p ≤ 4; "8c" HNF4α+ = 4 < p ≤ 8; ">8c" HNF4α+ = 8 < p.
        10. To reconstruct the spatial patterning of ploidy subgroups, separate the nuclear data within each sample spreadsheet according to the corresponding fields in which they were acquired. Then use associated nuclear x/y coordinates (from step 5.5) to plot ploidy subgroups in 2D (Figure 5C).

Subscription Required. Please recommend JoVE to your librarian.

Representative Results

or Start trial to access full content. Learn more about your institution’s access to JoVE content here

This method has been used to measure the impact of cholestatic injury on the adult mouse liver by feeding animals for 0−21 days with a hepatotoxic diet containing 0.1% 3,5-diethoxycarbonyl-1,4-dihydrocollidine (DDC)17. Chronic DDC feeding results in hepatocellular injury increased ploidy and periportal expansion of NPCs. The user should be aware that mouse strain and age-dependent differences may exist in nuclear ploidy and that all analyses have been performed using adult female C57BL/6 mice aged 12−16 weeks.

After HNF4α immunolabelling (protocol section 3), it is important to check all slides using conventional fluorescence microscopy, to ensure good quality fixation and staining (Figure 2A). Smearing or blurring of Hoechst can indicate inadequate fixation or sample degradation prior to fixation (Figure 2B), in which case return to protocol section 2 and shorten the time between sectioning and fixation (step 3.1). Successful immunolabelling with the HNF4α antibody can easily be judged at this stage by clear discrimination of positively labelled hepatocyte nuclei, typically larger and more rounded than those of NPCs (Figure 2A). Flattened/elliptical endothelial nuclei within the parenchyma, or dense patches of cells that expand in periportal areas following DDC injury can serve as a useful visual reference for identifying HNF4α- NPCs when assessing the success/failure of immunostaining.

Nuclear segregation and HNF4α threshold parameters (steps 5.2−5.4) should be carefully optimized prior to automatic image analysis (step 5.6) to broadly reflect the visual pattern of immunostaining observed by conventional fluorescence microscopy at the end of protocol section 3 (Figure 2C). Examples of optimal vs suboptimal nuclear segmentation and HNF4α threshold protocols are summarized in Figure 2D. Following image analysis (step 6.1.3), the data should reflect increasing numbers of NPCs in the liver with DDC injury (Figure 2E), from 52% ± 2.0% of nuclei in control livers to 72.8% ± 1.4% after 21 days of DDC treatment. Hepatocytes represent 48.0% ± 2.0% of total nuclei in control livers, concordant with previous analyses of liver histology showing that hepatocytes occupy 70−85% of the tissue volume, but only 45−50% of total liver cells18,19. A small but significant reduction in numbers of HNF4α+ nuclei is observed during the first 14 days of DDC feeding (Figure 2F). A frequency distribution plot of hepatocyte nuclear area (step 6.2) shows a peak HNF4α+ nuclear area in control livers in the 40−50 µm2 size range, and a clear right-shift in nuclear size after DDC injury (Figure 2G); consistent with increased ploidy and hepatocellular hypertrophy12.

In healthy (day 0) control livers, 63.4% ± 1.7% of HNF4α+ nuclei have a "simple" circular morphometry (Figure 5A). This figure decreases to 46.8% ± 5.7% (P = 0.042) after 21 days of DDC injury, reflecting increased complexity in nuclear morphometry presumably associated with shifting between ploidy states during polyploidization (see "Interpretation of nuclear morphometry" below). Representative examples of nuclear ploidy distributions obtained using this method in control liver sections are shown in Figure 5A, which describes how interpolation of DNA content allows stratification of individual cells within a single sample. Mean values for subsets of "complex" and "simple" HNF4α+ cells are also shown (Figure 5A). The data are consistent with previous estimates of polyploidy in 80−90% of adult murine hepatocytes2. The frequency of complex nuclei (36.6% ± 1.7%) in control livers also approximates to that of binuclear cells (35%)20, though data should be strictly regarded as a measure of nuclear rather than cellular ploidy (see "Interpretation of nuclear morphometry" below). Comparison of relative ploidy between control (day 0) and DDC treated groups should reflect significant loss of 2c and 4c hepatocyte nuclei with injury together with increased numbers of >8c cells (Figure 5B). Relative positional information for each ploidy subgroup can be interrogated by scatter plot of x-y coordinates associated with each nucleus within the dataset or by retrieving the 2D location of particular hepatocyte subsets within the high-content image analysis software (Figure 5C).

Calibration
To assess the validity of the NPC calibration method used, dual immunolabeling of liver sections was performed using antibodies to HNF4α and proliferative marker Ki-67 (Figure 6A,B). These data showed enrichment for Ki-67, which labels cells in all active phases of the cell cycle, on the right side of the NPC minimal DNA distribution curve (between S2c and S4c) - where NPCs would be expected to be replicating DNA and therefore have >2c ploidy (Figure 6A). After internal calibration of all control and injured liver samples studied, Ki-67, was significantly enriched (P < 0.0001) in "simple" NPC nuclei with an estimated ploidy of >2c (82.5% ± 6.6% SD, n = 12) compared to those with ≤2c ploidy (17.5% ± 6.6% SD, n = 12) (Figure 6B), indicating successful ploidy calibration. These data support the validity of the method used. Also, assuming accurate thresholding of Ki67, they provide some quantitative insight into the extent to which subequatorial nuclear masks from higher ploidy groups "contaminate" groups below.

To further test the validity of the NPC calibration method, an external calibrator was introduced based on the previously reported nuclear volume of mouse 2N hepatocytes (155.8 µm3)14. When this figure was used in combination with an average value of Hoechst intensity for HNF4a- nuclei the resulting estimate for mean hepatocyte ploidy in the control livers was indistinguishable from that of the internal (S2c) calibrator (Figure 6C). Moreover, estimates of mean hepatocyte ploidy in mice of the C57BL/6 mouse strain of comparable age, were also similar, confirming that prior empirical knowledge of 2N hepatocyte nuclear size is not required, making this internally controlled methodology for estimating nuclear ploidy fully autonomous.

Interpretation of nuclear morphometry
The method described provides a ploidy readout for hepatocyte nuclei with "simple" circular morphometry. Exclusion of "complex" nuclei is based on the hypothesis that they represent a proportion of binuclear hepatocytes with overlapping/touching nuclear masks, making accurate ploidy determination for this subset more challenging (Figure 7A). Importantly, segregation of nuclei according to circularity does not enable the user to distinguish between the nuclei of mononuclear hepatocytes and those of binuclear cells, in which two nuclei of similar ploidy are clearly separated within the cell. This was empirically tested by manually selecting binuclear and mononuclear cells from the image datasets and assessing their segregation by the algorithm (Figure 7B). Nuclei of binuclear hepatocytes that were physically close (Figure 7C) but "not touching" were categorized by the algorithm as "simple", whereas those that were "touching" were clearly discriminated as "complex". Hence, this assay does not provide a readout of cellular ploidy in the liver, given that nuclei of binuclear cells are subdivided between the "simple" and "complex" subclasses (see Discussion). However, some insight into the switching between states of cellular and nuclear ploidy may be gained from this data simply by plotting histograms of nuclear morphometry and nuclear size and applying a model of how "complex" and "simple" states are transitioned between during polyploidization (Figure 7D). In control livers three phases (I−III) of nuclear morphometry are clearly observed (Figure 7E). They represent the clustering of circular 2N (I), 4N (II) and 8N (III) nuclear masks respectively (as illustrated in Figure 7D). Mononuclear 16N hepatocytes are extremely rare in adult mouse livers16,18,21, hence the 16N cellular ploidy group is comprised almost entirely of binuclear cells with 8N nuclei, located within, and to the right, of phase III (Figure 7E), explaining the drop in circularity to the right of phase III. Interestingly, upon injury (DDC day 14), a quantitative shift towards increased complexity ("binuclearity") begins in phases I (reflecting 2n to 2x2n) and phases III (reflecting 8n to 2x8n), before it is finally consolidated in all three phases (I−III). The authors speculate that this shift towards increased complexity is due to an increased cellular ploidy resulting from stalled cytokinesis, whereas to the right of phase III the opposite trend is observed, due to increased representation of circular 16N mononuclear cells in the injured liver due to endoreplication. These observations will of course need to be tested by adapting the method to properly account for cellular ploidy (see Discussion).

Figure 1
Figure 1: Summary of workflow. Liver tissue is harvested (1), cryosectioned (2), fixed and immunolabelled with an HNF4α antibody allowing for parenchymal and non-parenchymal cells (NPCs) to be discriminated (3). Once processed, samples are digitized using a high-content imaging platform using automated image capture (4) and analysis (5). Cells are segmented by Hoechst nuclear fluorescence and HNF4α immunofluorescence thresholding. Next, Hoechst nuclear area ("A") and circularity ("C") are calculated. Finally, the data are analyzed (6); HNF4α- NPCs are quantified (i) and HNF4α+ hepatocyte nuclei are separated into two subsets ("simple" and "complex") according to nuclear circularity (ii). Interpolation of hepatocyte nuclear ploidy is then performed for all "simple" nuclei as a function of nuclear radius (r) and mean Hoechst fluorescence intensity (as a proxy for nuclear DNA density) (iii). The data are then stratified using NPCs as an internal 2N calibrator (iv) before compiling a sample summary (v). Please click here to view a larger version of this figure.

Figure 2
Figure 2: High-content image analysis and cytometric profiling of the mouse liver during chronic DDC feeding. (A) A representative confocal image of HNF4α/Hoechst immunofluorescence staining of the adult mouse liver after 21 days of feeding with a diet containing 0.1% DDC; image shows rounded HNF4α+ hepatocyte nuclei ("H") and expansion of HNF4α- NPCs in areas surrounding the portal vein ("PV"). (B) Examples of optimal ("correct") and suboptimal ("incorrect") nuclear Hoechst staining indicating poor fixation. (C) Use of high-throughput image analysis platform to segregate hepatocytes and NPCs according to nuclear Hoechst staining and HNF4α immunolabelling. Software masks (red/green lines) show how nuclei are correctly segmented according to Hoechst fluorescence and sorted into hepatocytes (+) or NPCs (-) according to HNF4α status. (D) A guide to optimizing the setup for segmentation/threshold analysis. Superimposed nuclear masks recognized by the software are indicated by green/blue lines for nuclear segmentation and green/blue (HNF4α+) or red/blue (HNF4α-) for threshold analysis (H = hepatocyte). Troubleshooting: Nuclear detection sensitivity set too low (i), or too high (ii). Threshold for HNF4α set too low (iii), or too high (iv). (E,F) Quantitative analysis of NPC and hepatocyte nuclei during DDC feeding: (E) HNF4α- and (F) HNF4α+ nuclear densities are compared against time of DDC treatment (days). A total of 5.7 x 105 cells were analyzed, from 4−6 animals per timepoint. Data are presented as mean + SEM. **P < 0.01 and ***P < 0.001. One-way ANOVA was used to compare means. Significance P values were calculated using Fisher's least significant difference (LSD) test. (G) Frequency distribution of HNF4α+ nuclear area during DDC treatment. The data show a right-shift in hepatocyte nuclear area during injury consistent with cellular hypertrophy and polyploidization. A total of 2.5 x 105HNF4α+ nuclei were analyzed, from 4−6 animals per timepoint. This figure has been modified from Manzano-Núñez et al.17. Please click here to view a larger version of this figure.

Figure 3
Figure 3: Automated analysis of hepatocyte nuclear ploidy using custom written software. (A) Screenshot showing correct formatting of spreadsheet data for input into the nuclear ploidy analysis software. Columns containing essential data (step 5.5 of the protocol) are highlighted yellow. All column titles should precisely match those indicated. (B) Screenshot showing how individual spreadsheet files containing data from biological replicates ("Sample1", "Sample2", etc.) should be named and organized in subfolders for each condition (entitled "Control-d0" and "Injured-d14" in this example). (C) Screenshot after successful installation of the ploidy application (red circle). When the application is launched (by clicking "Ploidy_Appl..") in the MY APPS tab of the toolstrip the "Ploidy_GUI" appears (lower panel). The experiment name ("Sample") and paths to the control (e.g., "Control-d0") and test (e.g., "Injured-d14") datasets are entered before clicking Run. The software then calculates, calibrates and stratifies nuclear ploidy for all samples using the "Control-d0" dataset to generate thresholds for minimal DNA content. (D) Data output from Ploidy_Application shows individual data files automatically saved in each sample folder (i) containing absolute and percentage numbers of "simple" nuclei in each ploidy group. For each condition (in this case both "Control-d0" and "Injured-d14"), a summary folder is also automatically generated containing mean nuclear ploidy estimates for all "simple" hepatocyte and non-hepatocyte nuclei (ii) and a breakdown of how nuclear ploidy is stratified for each sample (iii). Please click here to view a larger version of this figure.

Figure 4
Figure 4: The use of NPCs as an internal ploidy calibrator. (A) Graph showing the impact of DDC injury on mean minimal DNA content (m) of hepatocyte (HNF4α+) and NPC (HNF4α-) nuclei. All data are normalized to day 0 NPCs (n = 4 animals per timepoint). (B) Histogram describing the distribution of NPCm values in a single representative liver sample (day 0, total of 7,180 nuclei). The schematic (above) shows how circular NPC masks can derive from cells with a 2−4c DNA content. The aim of the calibration method is to define the stratification threshold representing 4c (S4c) at the upper limit of the NPCm distribution (dotted line), while minimizing noise due to segmentation errors at the extremes of the distribution curve. (C) Changes in mean Hoechst intensity and nuclear area for NPC (HNF4α-) nuclei are plotted. To avoid segmentation error only those nuclei with a corresponding NPCm value of within 1 SD of the mode NPCm value are scrutinized (yellow box). Within this range the 2c−4c transition size (t) is calculated and used as an anchor point within the data to estimate the S4c. Please click here to view a larger version of this figure.

Figure 5
Figure 5: High-throughput in situ analysis of nuclear ploidy in the mouse liver during chronic DDC feeding. (A) Analysis of control adult liver using the described methodology. HNF4α+ hepatocyte nuclei from 2D liver sections are subdivided according to Hoechst nuclear circularity into two groups: "simple" and "complex". (Top) Representative fluorescence Hoechst images of cells belonging to these two groups are shown. (Left) Scatterplot showing stratification of simple HNF4α+ nuclei from one sample (day 0) according to interpolated ploidy value, nuclear area and mean nuclear Hoechst intensity. (Right) Pie chart detailing the typical breakdown of HNF4α+ cells in control liver (day 0) indicating the proportions of each nuclear ploidy subclass. A total of 6.7 x 104 HNF4α+ nuclei from 4 animals were analyzed. (B) Impact of DDC liver injury on hepatocyte nuclear ploidy by high-throughput in situ analysis. Graphs demonstrate the relative decrease in the proportion of 2c and 4c hepatocyte nuclei within the first 14 days of DDC feeding while >8c polyploid nuclei dramatically increase in number. A total of 1.5 x 105 HNF4α+ nuclei were analyzed (n = 4 animals per timepoint). Data are presented as mean + SEM. **P < 0.01 and ***P < 0.001. One-way ANOVA was used to compare means. Significance P values were calculated using Tukey's multiple comparison test. (C) Example to show how nuclear ploidy subclasses can be spatially tracked within the parenchyma using this method, by interrogating high-content imaging data with the same quantitative criteria used for ploidy stratification (circularity, nuclear size and mean Hoechst intensity). Hoechst fluorescence images are shown with software masks (red dots) marking 2c nuclei in the liver at two timepoints during chronic DDC feeding (day 14 and 21). Portal vein (blue dotted line) and periportal areas in which NPCs expand (yellow line) are indicated. Please click here to view a larger version of this figure.

Figure 6
Figure 6: Critical assessment of the NPC calibration method. (A,B) Proliferating NPCs are successfully categorized with a >2c ploidy score. (A) Histogram of NPC minimal DNA content from control livers immunolabeled with antibodies to HNF4α and proliferative marker Ki-67 (n = 4, data are presented as mean + SEM). Stratification thresholds for 2c (S2c) and 4c (S4c) are indicated. (B) Stratification of NPCs according to the described methodology results in a significant enrichment of Ki-67 immunolabelling in nuclei assigned a >2c ploidy score (n = 12). Data are presented as mean + SEM. Unpaired t test was used to compare the means ****P < 0.0001. (C) External validation of the NPC calibration method. Estimates of mean hepatocyte nuclear ploidy obtained using the internal NPC calibrator method were compared to those obtained by calibration of the same samples (Control C57BL/6 mouse liver 3−4 months, n = 4) with a known nuclear volume for 2N hepatocytes14. Data is also presented from two independent analyses21,22 describing hepatocyte nuclear ploidy from mice of the same strain at ages 2−6 months (shown to the right of the dotted line). Please click here to view a larger version of this figure.

Figure 7
Figure 7: Hepatocyte nuclear morphometry has a "complex" relationship with cellular ploidy. (A) Summary of how hepatocyte cellular ploidy (2N, 4N, 8N and 16N) is partially segregated by 2D analysis of nuclear morphometry. Binuclear cells (red) are subdivided between "simple" ("S") and "complex" ("C") morphometries depending on whether nuclei appear to be touching or not. (B,C) Individual hepatocytes were manually selected and analyzed for nuclear morphometry and internuclear spacing. (B) Binuclear hepatocytes with "touching" nuclei were "complex" (100% ≤0.8), whereas mononuclear cells (black) and binuclear hepatocytes with non-touching nuclear masks were "simple" (94% >0.8). (C) "Simple" nuclei of binuclear hepatocytes could be distinguished from those of mononuclear cells on account of significantly reduced inter-nuclear spacing (n = 3, total of 94 nuclei analyzed). (D) Model approximating how cellular ploidy states from panel A might be distributed in terms of 2D nuclear morphometry and nuclear area resulting in clustering of simple circular forms into four phases (I−IV). (E) Comparison of HNF4α+ nuclear morphometry/size plots in control livers (day 0) and after 14 days (left) and 21 days (right) of DDC injury. Morphometry phases are indicated above (I−V). Arrows indicate shifts in nuclear morphometry resulting from injury that are consistent with binuclearization of 2c ("a"), 4c ("b") and 8c ("c") nuclei, together with increased mononuclearization of the 16N cellular ploidy class (d). Total of 29−30 x 103 nuclei analyzed per condition (n = 2). Please click here to view a larger version of this figure.

Supplemental Files: Demonstration Datasets. Please click here to view these files.

Subscription Required. Please recommend JoVE to your librarian.

Discussion

or Start trial to access full content. Learn more about your institution’s access to JoVE content here

A high-content, high-throughput approach for the analysis of tissue remodeling and estimation of hepatocyte nuclear ploidy in the murine liver is described. Once familiar with the procedure, a user can process, image and analyze multiple samples in a 3−5 day period, generating large testable datasets that provide a detailed signature of liver health. Given the simplicity of the sample preparation method, together with the large numbers of cells and tissue area analyzed (on average 14 mm2/sample), results are robust and highly reproducible. Automation of image capture and analysis also removes user error and potential bias from these important steps. An important innovation is the use of NPCs as an internal ploidy calibrator that enables relative assessment of hepatocyte nuclear DNA content both within and between samples. Incorporation of an HNF4α labelling step is therefore key to providing this protocol with a unique technical advantage compared to previously published 2D methods3,12,22. In contrast, the relative simplicity of the methodology in comparison with 3D reconstruction workflows18 makes it technically less laborious and potentially more flexible.

Compared to the precision method of flow cytometry, an important caveat to extrapolating nuclear DNA content from 2D tissue sections, is the limited confidence that can be attributed to the categorization of individual nuclei with regards to ploidy status. Added to this is the inherent bias within SIA based approaches to overrepresent smaller ploidy subgroups due to subequatorial sampling. However, by normalizing data to an internal standard and taking a large population based approach, the error due to these effects is mitigated and comparable across samples. Hepatocytes are characterized by a highly rounded nuclear morphology, compared to for example NPCs, meaning they are particularly amenable to accurate estimation of DNA content based on nuclear cross-sectional area alone10,11,14,15,23. The SIA-based approach has been refined in this protocol to account both for nuclear circularity and DNA density by integrating measures of morphometry and mean Hoechst fluorescence intensity resulting in an estimated "minimal DNA content" descriptor for individual nuclei. Importantly, the use of NPCs as a 2−4N ploidy control provides an important internal standard for objective calibration and stratification of minimal nuclear DNA content, making the methodology described applicable to samples of any species, or format, given that an appropriate antibody for HNF4α (or similar hepatocyte nuclear marker) can be sourced.

Although assessment of nuclear ploidy has been shown to provide useful signatures for liver disease progression10,13, in order to fully ascertain the diversity of ploidy changes within the liver it would be both desirable and necessary to adapt the described methodology to account for hepatocellular perimeter and thus cellular ploidy. Mapping of cellular ploidy has previously been achieved by labelling of the hepatocyte perimeter using markers such as beta catenin10,13,24, actin12,22 and cytokeratin10,11 in human and mouse liver samples. However, when this was tested after DDC injury, dramatic epithelial remodeling precluded reliable assessment of hepatocellular perimeter both by phalloidin (data not shown) or antibodies to beta catenin17. Hence, whilst this approach is feasible, it may not be applicable to all injury models, but if achieved would advance mapping of cellular ploidy as well as making estimates of hepatocyte size and number more accurate. It also remains plausible that by accounting for additional nuclear parameters, such as internuclear spacing (Figure 7C), mononuclear cells could be discriminated from "simple" binuclear hepatocytes, and that further segregation of "complex" binuclear cells could be achieved by radial measurements of the nuclei that their 2D masks contain.

Given that validated human HNF4α antibodies exist for FFPE tissue25 and that internal calibration frees this methodology of any species-specific limitations, the protocol is almost immediately applicable to human samples. Thus, it has considerable potential to provide a benchmark for high-throughput analysis of hepatocyte nuclear ploidy and liver injury in human disease. Also, by multiplexing with other antibodies, this method can reveal new roles for particular hepatocyte subsets and their response to liver injury and disease. To this end, we have successfully combined the methodology with immunostaining for the proliferative nuclear marker Ki-67 (Figure 6), which enables useful information to be gleaned - including identification of non-proliferating 2N populations of NPCs for improved internal calibration of ploidy (Noon, unpublished data 2019). Hence, by coupling flexibility with the positional and quantitative data that the method provides, we suggest that its future applications will improve understanding of the role of polyploidy in the liver.

Subscription Required. Please recommend JoVE to your librarian.

Disclosures

The authors have nothing to disclose.

Acknowledgments

This work was funded by the Spanish MINECO Government grants BFU2014-58686-P (LAN) and SAF-2017-84708-R (DJB). LAN was supported by a national MINECO Ramón y Cajal Fellowship RYC-2012-11700 and Plan GenT award (Comunitat Valenciana, CDEI-05/20-C), and FMN by a regional ValI+D studentship of the Valencian Generalitat ACIF/2016/020. RP would like to acknowledge Prof. Ewa K. Paluch for funding. We thank Dr. Alicia Martínez-Romero (CIPF Cytometry service) for help with the IN Cell Analyzer platform.

Materials

Name Company Catalog Number Comments
3,5-diethoxycarboxynl-1,4-dihydrocollidine diet (DDC) TestDiet 1810704 Modified LabDiet mouse diet 5015 with 0.1% DDC
Alexa Fluor 488 donkey anti-goat IgG (H+L) Invitrogen A11055 Dilution 1:500
Bovine Serum Albumin Sigma-Aldrich A7906
Cryostat Leica CM1850 UV Leica biosystems CM1850 UV Tissue sectioning
Fluorescent Mounting medium Dako S3023
GraphPad Prism GraphPad Software Prism 8 Statistical software for graphing data
Hoechst 33342 Sigma-Aldrich B2261 Final concentration 5 µg/mL
IN Cell Analyzer 1000 GE Healthcare Bio-Sciences Corp High-Content Cellular Imaging and Analysis System
MATLAB MathWorks R2019a Data analytics software for automated analysis of nuclear ploidy
Microscope coverslides VWR International 630-2864 Size of 24 x 60 mm
Microsoft Office Excel Microsoft Speadsheet software
OCT Tissue Tek Pascual y Furió 4583
Paraformaldehyde Panreac AppliChem 141451.121
Pen for immunostaining Sigma-Aldrich Z377821-1EA 5mm tip width
Polysine Microscope Slides VWR International 631-0107
Rabbit polyclonal Anti-HNF4α Thermo Fisher Scientific PA5-79380 Dilution 1:250 (alternative)
Rabit polyclonal Anti-HNF4α Santa Cruz Biotechnology sc-6556 Dilution 1:200 (antibody used in the study)
Tween 20 Sigma-Aldrich P5927

DOWNLOAD MATERIALS LIST

References

  1. Gentric, G., Desdouets, C. Polyploidization in liver tissue. American Journal of Pathology. 184, (2), 322-331 (2014).
  2. Duncan, A. W., et al. The ploidy conveyor of mature hepatocytes as a source of genetic variation. Nature. 467, (7316), 707-710 (2010).
  3. Gentric, G., Desdouets, C. Liver polyploidy: Dr Jekyll or Mr Hide? Oncotarget. 6, (11), 8430-8431 (2015).
  4. Wilkinson, P. D., et al. The Polyploid State Restricts Hepatocyte Proliferation and Liver Regeneration in Mice. Hepatology. 69, (3), 1242-1258 (2019).
  5. Wilkinson, P. D., et al. Polyploid Hepatocytes Facilitate Adaptation and Regeneration to Chronic Liver Injury. The American Journal of Pathology. 189, (6), 1241-1255 (2019).
  6. Zhang, S., et al. The Polyploid State Plays a Tumor-Suppressive Role in the Liver. Developmental Cell. 44, (4), 447-459 (2018).
  7. Chao, H. W., et al. Circadian clock regulates hepatic polyploidy by modulating Mkp1-Erk1/2 signaling pathway. Nature Communications. 8, (1), 2238 (2017).
  8. Celton-Morizur, S., Merlen, G., Couton, D., Margall-Ducos, G., Desdouets, C. The insulin/Akt pathway controls a specific cell division program that leads to generation of binucleated tetraploid liver cells in rodents. Journal of Clinical Investigation. 119, (7), 1880-1887 (2009).
  9. Wang, M. J., Chen, F., Lau, J. T. Y., Hu, Y. P. Hepatocyte polyploidization and its association with pathophysiological processes. Cell Death & Disease. 8, (5), e2805 (2017).
  10. Gentric, G., et al. Oxidative stress promotes pathologic polyploidization in nonalcoholic fatty liver disease. Journal of Clinical Investigation. 125, (3), 981-992 (2015).
  11. Toyoda, H. Changes to hepatocyte ploidy and binuclearity profiles during human chronic viral hepatitis. Gut. 54, (2), 297-302 (2005).
  12. Miyaoka, Y., et al. Hypertrophy and Unconventional Cell Division of Hepatocytes Underlie Liver Regeneration. Current Biology. 22, (13), 1166-1175 (2012).
  13. Bou-Nader, M., et al. Polyploidy spectrum: a new marker in HCC classification. Gut. (2019).
  14. Danielsen, H., Lindmo, T., Reith, A. A method for determining ploidy distributions in liver tissue by stereological analysis of nuclear size calibrated by flow cytometric DNA analysis. Cytometry. 7, (5), 475-480 (1986).
  15. Guidotti, J. E., et al. Liver Cell Polyploidization: A Pivotal Role for Binuclear Hepatocytes. Journal of Biological Chemistry. 278, (21), 19095-19101 (2003).
  16. Severin, E., Meier, E. M., Willers, R. Flow cytometric analysis of mouse hepatocyte ploidy - I. Preparative and mathematical protocol. Cell and Tissue Research. 238, (3), 643-647 (1984).
  17. Manzano-Núñez, F., et al. Insulin resistance disrupts epithelial repair and niche-progenitor Fgf signaling during chronic liver injury. PLoS Biology. 17, (1), e2006972 (2019).
  18. Morales-Navarrete, H., et al. A versatile pipeline for the multi-scale digital reconstruction and quantitative analysis of 3D tissue architecture. eLife. 4, e11214 (2015).
  19. Baratta, J. L., et al. Cellular organization of normal mouse liver: A histological, quantitative immunocytochemical, and fine structural analysis. Histochemistry and Cell Biology. 131, (6), 713-726 (2009).
  20. Pandit, S. K., et al. E2F8 is essential for polyploidization in mammalian cells. Nature Cell Biology. 14, (11), 1181-1191 (2012).
  21. Vinogradov, A. E., Anatskaya, O. V., Kudryavtsev, B. N. Relationship of hepatocyte ploidy levels with body size and growth rate in mammals. Genome. 44, (3), 350-360 (2001).
  22. Tanami, S., et al. Dynamic zonation of liver polyploidy. Cell and Tissue Research. 368, (2), 405-410 (2017).
  23. Kudryavtsev, B. N., Kudryavtseva, M. V., Sakuta, G. A., Stein, G. I. Human hepatocyte polyploidization kinetics in the course of life cycle. Virchows Archiv B Cell Pathology Including Molecular Pathology. 64, (1), 387-393 (1993).
  24. Gentric, G., Celton-Morizur, S., Desdouets, C. Polyploidy and liver proliferation. Clinics and Research in Hepatology and Gastroenterology. 36, (1), 29-34 (2012).
  25. Uhlén, M., et al. Tissue-based map of the human proteome. Science. 347, (6220), 1260419 (2015).
A High-Throughput In Situ Method for Estimation of Hepatocyte Nuclear Ploidy in Mice
Play Video
PDF DOI DOWNLOAD MATERIALS LIST

Cite this Article

Manzano-Núñez, F., Peters, R., Burks, D. J., Noon, L. A. A High-Throughput In Situ Method for Estimation of Hepatocyte Nuclear Ploidy in Mice. J. Vis. Exp. (158), e60095, doi:10.3791/60095 (2020).More

Manzano-Núñez, F., Peters, R., Burks, D. J., Noon, L. A. A High-Throughput In Situ Method for Estimation of Hepatocyte Nuclear Ploidy in Mice. J. Vis. Exp. (158), e60095, doi:10.3791/60095 (2020).

Less
Copy Citation Download Citation Reprints and Permissions
View Video

Get cutting-edge science videos from JoVE sent straight to your inbox every month.

Waiting X
simple hit counter