**Background:** Studies in the digestive tract often required precision quantification of intestinal volume to observe the effect of certain intervention/condition. Application of stereological methods could bring unbiased and accurate results but commercially computer-assisted systems are not widely available. ImageJ-FIJI is an open source software, which could become an alternative choice in the stereological measurement process. **Aim:** This study describes simple stereological quantification methods during volume estimation of jejunum-ileum intestinal layers of the rats using a light microscope and ImageJ-FIJI stereological tool. **Material and Methods:** Six 3-months old male Sprague-Dawley rats were terminated and jejunum-ileum was harvested after perfusion. After removal of intestinal luminal content, whole jejunum-ileum weight was measured. The organ was sampled as 6-10 slabs of 1 cm length in a systematic uniformed random sampling manner. Slabs were cut longitudinally at random angles before flattened and put on filter papers for subsequent tissue processing into 2-3 paraffin blocks. One section of 3 μm thick was sampled from each block, stained using toluidine blue and documented using a light microscope connected to a microstepper apparatus. The volume of the intestinal layers was estimated using a point-counting grid on Image J-FIJI software. **Result:** We compared two sets of counting methods i.e. minimal counting (MC) and rigorous counting (RC) approaches that differ in their respective a/p value. Quantification using RC approach resulted in significantly higher estimated volume of tunica submucosa and tunica muscularis while having more preferable stereological accuracy parameters (CE<5% & CV<10%). **Conclusion:** Although it required longer counting time, rigorous approaches resulted in higher accuracy while still within the range of rule of thumb criteria of 0.2 < CE^{2}/CV^{2} < 0.5.
**Keywords:** Digestive tracts, ImageJ-FIJI, quantification, stereology, volume
**How to cite this URL:** Wicaksono SA, Sumiwi YA, Paramita DK, Susilowati R. ImageJ-FIJI-Assisted estimation of intestinal layers' volume: Study in Jejunum–Ileum of rats. J Microsc Ultrastruct [Epub ahead of print] [cited 2023 Feb 8]. Available from: https://www.jmau.org/preprintarticle.asp?id=362489 |
Introduction | | |
Jejunum–ileum is a segment of the small intestine that is important in the digestion and absorption of nutrients. Studies using laboratory animal models such as rats are widely used to provide preclinical data for studying the pathophysiology of diseases and the development of therapy.^{[1],[2]} Various studies have reported the effect of genetic variation, environmental exposure, or pharmacological interventions on jejunum–ileum histomorphology, reporting quantitative parameters such as the height of intestinal villi, depth of the crypt, the thickness of the layer as well as density values of several structure or cells.^{[3],[4],[5]} In this study, we will focus on the layers of jejunum–ileum, i.e. tunica mucosa, tunica submucosa, tunica muscularis, and tunica serosa. The layers may undergo volume alteration due to pathological processes, for example, infection, inflammation, and degenerative diseases.^{[6],[7]} Therefore, quantitative data on the volume of these layers may become important parameters for certain studies.
Histological sections are two-dimensional thin sections of three-dimensional organs. Counting on sections sampled using simple random methods may result in sampling bias, while measurement with inappropriate probe may introduce geometrical bias. Furthermore, The data are often reported as density values without total data of the whole organ. These limitations were associated with a high risk of type I and II error, limited sensitivity, and risk of confounding interference on the statistical accuracy of the study.^{[8]}
The unbiased stereological techniques ensure analyzing efficiency without losing precision in quantification studies on histological sections.^{[9]} Stereology quantification not only analyzes three-dimensional (3-D) structures through 3-D reconstruction using stochastic geometry theory, but they also have strong statistical and mathematical foundations which are used to prevent assumptions or bias during the analytical process.^{[8],[10]} Some efforts to perform unbiased methods in the quantification of the volume of jejunum–ileum layers of rodents have been reported^{[7],[11]} using planimetry or lineal integration. However, the point-counting method using a zero-dimensional probe has shown superior compared to the two earlier mentioned methods.^{[9]} A stereological study conducted before the 1990s was done by superimposing point grids on transparent plastic on top of printed micrographs which is tiring and time-consuming.^{[5],[7],[11]} Establishment of computer-assisted and commercially available stereological quantification systems such as the newCAST (Visiopharm, Denmark) and Stereo Investigator (MBF Bioscience, Vermont, USA) has greatly improved the efficiency of the process.^{[12]} Those systems massively aid the stereological quantification processes but only can be procured by well-funded laboratories due to their high cost.
Fiji Is Just ImageJ (FIJI) is a variant of popularly adopted software ImageJ with additional plugins that are suitable for the stereological procedure. It is bundled with many plug-ins, including TrakEM2, which is full of stereology-related plugins.^{[12]} During volume estimation, the software could generate a virtual grid with an easily adjusted interpoint distance that is notated on the area per point (a/p) value. Such flexibility could help researchers to achieve optimum procedures based on the size of the specimen. To cope with the limitation, researchers developed open-source alternative software to visualize, inspect, quantify, and validate image data in the form of ImageJ software National Institute of Health, Bethesda, Maryland, USA (NIH). Due to its open-source characteristics, ImageJ gains tremendous growth and enhancement from collaborative user and developer communities worldwide.
Optimum stereological quantification requires observation in 8-10 histological sections obtained in a systematic uniform random sampling manner, resulting in 100–200 counts with the coefficient of error (CE) <5%, coefficient of variation (CV) <10%, and fulfilling the “rule of thumb” 0.2 < CE^{2}/CV^{2} < 0.5.^{[8],[13]} In this study, we use the term minimum counting approach for the measurement that meets those criteria. However, sometimes obtaining 100–200 counts does not end with desirable precision. Therefore, a more rigorous criterion using 600–900 counts has been suggested to be the criteria for optimum stereological estimates.^{[14]} In this study, we perform two sets of point-counting approaches, i.e. minimum counting and rigorous approaches in obtaining volume estimation of the jejunum–ileum layers of rats assisted by ImageJ-FIJI software.
Materials and Methods | | |
**Animal and ethical consideration**
Six 3-month-old male Sprague-Dawley rats, weighing 150–200 g were obtained from the Veterinary Faculty of Institut Pertanian Bogor (IPB, Bogor, Indonesia). The study protocol was approved by the Medical and Health Research Ethics Committee of the Faculty of Medicine, Public Health and Nursing Universitas Gadjah Mada (Ref No. KE/FK/0897/EC/2019, approved on August 1^{st} 2019) in conjunction of the Helsinki Declaration guidelines. The rats were euthanized under ketamine 100 mg/kg BW i.p., followed by transcardial perfusion using cold phosphate buffered-saline (PBS) pH 7*.*^{[15]}
The whole segment of jejunoileum was harvested and separated from the duodenum at the flexural duodenolienalis. It is separated from the colon at the ileocecal junction. The jejunum was not separated from the ileum since there are no distinct visible edge markers. After the intestinal lumen was cleansed using a glass cylinder, organ weight was measured and recorded. The organs were put on plain paper with a ruler on the side, and documented using a desk camera (Optilab, Miconos, Indonesia) for length measurement [Figure 1]. | Figure 1: Tissue processing. The whole segment was measured (a) then cut into 6-10 slabs in 1 cm length and sampled using SURS method (b). After being fixed and dehydrated, every 3-4 slabs were embedded into 1 paraffin block before sectioning using a rotary microtome in 3 μm thickness (c). Single tissue sections obtained from each block were stained with toluidine blue
**Click here to view** |
**Histological slide preparation and shrinkage estimation**
Jejunum–ileum was divided into slabs of 1 cm length and sampled using a systematic uniform random sampling scheme to obtain six to ten slabs per organ [Figure 1]a and [Figure 1]b.^{[16]} The slabs were cut longitudinally at random angles, flattened, and sandwiched between filter papers before being immersed in 4% paraformaldehyde in PBS pH 7. Following dehydration and clearing processes, three to four slabs were embedded together into one block with an orientation that facilitate longitudinal sectioning.^{[17]} The longitudinal orientation enables us to get a better observation of the myenteric plexus which is important for the main study. Paraffin blocks were sectioned using a Leica RM2235 rotary microtome (Leica Microsystems, Singapore) at 3-μm thick [Figure 1]c. Slides were processed and stained with toluidine blue before being dehydrated in a series of ethanol and mounted in a synthetic resin mounting medium.^{[18]} One section per block was used in subsequent measurements.
Tissue shrinkage level was calculated by comparing surface areas documented before dehydration and before embedding four random sections per rat using point-counting procedure. Quantification was conducted using the following formula.^{[12],[17]}
**Estimation of Jejunoileum layers volume**
Micrographs were obtained using a ×10 objective lens of a CX-23 light microscope (Olympus, Japan) connected to Optilab® Microstepper hardware (Miconos, Indonesia) that captures serial 60–100 photographs and stitches them into a single composite figure using Image Composite Editor software*.* The first figures in each section were scaled using Image Raster software (Miconos, Indonesia).
Each layer of jejunum–ileum was identified, and volume estimation was performed by a point-counting method using superimposed point grids generated by ImageJ-FIJI (NIH). A number of points that hit the area of interest (mucosa, submucosa, and muscularis) were divided by the total points which hit the whole area of the section, resulting in the volume fraction of each layer. In this study, two quantification approaches were compared in terms of measurement accuracy and precision which are reflected in CE and CV values. The first approach was minimum counting stereology calculation which was conducted by achieving 100–200 measurements from six to ten sections per animal.^{[19]} To achieve those targets, the volume fraction of each layer was calculated using an area per point of 75,000 μm^{2} for the whole jejunum–ileum, 50,000 μm^{2} on tunica mucosa, 10,000 μm^{2} on tunica submucosa, and 15,000 μm^{2} on tunica muscularis of jejunum–ileum, respectively. The second approach was a rigorous approach which target 600–900 measurements per animal^{[14]} which was achieved using an area per point of 15,000 μm^{2} for the whole jejunum–ileum, 10,000 μm^{2} on tunica mucosa, 2000 μm^{2} on tunica submucosa, and 3000 μm^{2} on tunica muscularis [Figure 2]. | Figure 2: Virtual grids with respective a/*P* values are shown: (a) 75,000 μm^{2}; (b) 50,000 μm^{2}; (c) 10,000 μm^{2}; and (d) 15,000 μm^{2} used for minimal counting stereological approach, whereas grids with a/*P* value of (e) 15,000 μm^{2}; (f) 10,000 μm^{2}; (g) 2000 μm^{2}; and (h) 3000 μm^{2} were used in rigorous stereological approach. Counted points that hit the structure of interest were marked by yellow circles
**Click here to view** |
The weight of jejunum–ileum was used as a proxy of the jejunum–ileum fresh-tissue volume based on the assumption that 1 mg of organ weight ≈ 1 mm^{3} of organ volume and specific gravity is equal in all layers of jejunum–ileum.^{[17]} The final volume of tunica mucosa, submucosa, and muscularis was calculated by multiplication of each layer's volume fraction with the fresh-tissue volume and shrinkage-corrected jejunum–ileum volume.^{[16]} The equation for the tunica muscularis volume estimation method is shown below as an example. Similar equations were used to estimate the volume of tunica mucosa and tunica submucosa.
The corrected volume of jejunum–ileum = Primary volume × (shrinkage correction)
V_{(m/sm/mm)} = Vv_{tm} × corrected volume of jejunoileum
Note:
Vv_{(m}/_{sm/mm)} = volume fraction of each intestinal layers (mucosa/submucosa/muscularis)
∑P_{(m}/_{sm/mm)} = total point counted in each intestinal layers (mucosa/submucosa/muscularis)
a/p_{(m}/_{sm/mm)} = value of area per point for each intestinal layer (mucosa/submucosa/muscularis)
∑P_{(total))} = total point counted in total intestinal layers
a/p_{(total)} = value of a/p for in total intestinal layers
*V*_{(m}/_{sm/mm)} = volume of each intestinal layers (mucosa/submucosa/muscular
**Stereological efficiency and accuracy**
The accuracy of the stereological procedure was measured by CE which was calculated using a formula described previously.^{[20],[21]} Value of CE <5%, CE ½ of observed CV, and fulfilling “rule of thumb” 0.2 < CE^{2}/CV^{2} <0.5 were considered optimum.^{[8]}
**Statistical analysis**
For normal data distribution, significant differences between estimated intestinal layer volume quantified using efficient and rigorous methods were analyzed using a student *t*-test on Prism 9.0 software (GraphPad, CA, USA) and would be presented in means ± standard deviation. Statistical differences were determined when *P* < 0.05.
Results | | |
The jejunoileum of the rats weighed 3998–4624 mg with 404–602 mm in length. The organ weight-to-body weight ratio ranges from 1.2% to 1.9%. The tissue processing into the paraffin block induced 44% tissue shrinkage, which resulted in 2092.97–2619.72 mm^{3} shrinkage volume with a CV of 6%. Under the microscope, the jejunum–ileum tissue stained with toluidine blue showed tunica mucosa, submucosa, and muscularis, while tunica serosa only appeared on some sections due to damage during tissue processing. Volume estimations were conducted on tunica mucosa, tunica submucosa, tunica muscularis, and whole jejunoileal area from 7 to 10 sections per rat with an average of 9 ± 1 sections. Using the minimum counting approach, we got counted points from 107 to 175, which is within 100–200 optimum criteria. CE of the measurements using the minimum counting approach was more than 5% but still <10%. However, the CE^{2}/CV^{2} were higher than 0.5. The rigorous approach resulted in 614–812 counted points, which is slightly under the desirable range of 700–1000. The CE of all of the measurements using the rigorous approach was <5%, and the CE^{2}/CV^{2} was within the optimum range of 0.2–0.5. Stereological parameters related to counting approaches and their accuracy and precisions are shown in [Table 1]. | Table 1: Stereological parameters to assess the accuracy and precision of the quantification based on minimum counting and rigorous approach
**Click here to view** |
The estimated volume of the jejunoileum layers is shown in [Figure 3]. The estimated volume measured with the minimum counting approach were 1811.13 ± 171.18 mm^{3}, 220.57 ± 29.18 mm^{3}, 318.54 ± 27.12 mm^{3}, respectively, for tunica mucosa, tunica submucosa, and tunica muscularis. A similar value for the volume of tunica mucosa (1719.28 ± 136.59 mm^{3}) was estimated using the rigorous approach with no significant difference from the result of the minimum counting approach. However, the rigorous approach resulted in a higher volume of tunica submucosa (270.53 ± 25.01 mm^{3}) and tunica muscularis (390.14 ± 22.77 mm^{3}; *P* < 0.05). On average, estimating the volume of the jejunoileum layers using the minimum counting approach required 1–2 h/organ, whereas using the rigorous approach required 5–6 h. | Figure 3: Comparison between the volume of each intestinal layer estimated using MC and RC approach. ***t*-test; *P* < 0.01, ^{ns}Not significant. MC: Minimum counting, RC: Rigorous counting
**Click here to view** |
Discussion | | |
This study provides the first direct comparison between two approaches to the points counting method, i.e. the minimum counting approach and the rigorous approach, for volume estimation of the jejunoileal layers of the rats. In this study, the mean volume of the tunica mucosa obtained from both approaches has no significant difference. The volume estimates of the tunica submucosa and tunica muscularis obtained by a rigorous approach are higher. For tunica muscularis, the mean volume is more comparable to the previous report.^{[7]} It produces a better CE at the expense of a longer duration of the counting process. Although it is more time-consuming, the CE^{2}/CV^{2} of the rigorous approach are 0.2–0.4, the range is still within the accepted value.^{[20]} On the other hand, the minimum counting approach provides a shorter counting time at the expense of higher CE values which are more than half of their respective CV. Moreover, all of the CE^{2}/CV^{2} results from the minimum counting approach are larger than 0.5, which was not ideal according to the rule of thumb for stereological data.^{[22]}
The CV^{2} value is affected by the CE^{2} of the estimation and the biological variance of the subjects. Increasing the number of animals is the best thing to do if the biological variance of the sample is high. When high CE^{2} became the major contributor to high CV^{2}, the most efficient way to improve the precision of the estimation is by increasing the counting.^{[23]} The rigorous approach is based on computer simulation data that recommends 700–1000 measurements to be conducted whenever it is possible to achieve a CE value of 5%.^{[14]} The volume estimation in the existing samples aiming to the rigorous counting criteria shows appropriate CE and CV. The rigorous approach is still more resource efficient than increasing the number of animals.
In [Table 2], the results from the rigorous counting method are compared with the data of previous reports. The volume of the tunica mucosa in this study is comparable to the result of Uribe *et al.*^{[24]} but lower than the estimation of Mayhew and Carson^{[7]} that perform the measurement in bigger rats with longer ileum–jejunum segments. The difference in the size of the organ may explain the volume difference in tunica submucosa and tunica muscularis.^{[7]} Furthermore, those previous reports.^{[7],[24]} performed point-counting method using printed micrographs superimposed by grids. The reports have no explanation of the grid calibration method and no mention of the value of the area per point. The vulnerability of the loose connective tissue forming tunica submucosa to tissue deformation due to the processing method may contribute to the much smaller volume of the tunica submucosa obtained in this study compared to the previous report.^{[7]} | Table 2: Subject characteristics summary, absolute and layer volume of tunica mucosa, tunica submucosa, and tunica muscularis of rodent's jejunum-ileum relative to body weight using a rigorous approach
**Click here to view** |
One limitation of our study was the measurement of the organ reference volume, which is determined from the assumption of 1 mg of organ weight equals 1 mm^{3} organ volume.^{[17]} The water immersion method, according to Archimedes' principle provides an easy alternative to measure organ volume but has limited application in measuring small organs such as organs of small rodents and is not compatible with our subsequent tissue processing method. The current most accurate method in organ volume measurement is computed tomography scan or magnetic resonance imaging, but those methods are currently not widely used due to their high cost.^{[25],[26]}
Conclusions | | |
The present study demonstrated a comparison between two approaches in the stereological estimation of the volume of jejunum–ileum layers using images obtained from light microscopy assisted by ImageJ-FIJI software. This study recommended the application of a rigorous approach with a target of 600–900 measurements. Although the approach requires a longer counting time, it may achieve optimum CE, CV, and CE^{2}/CV^{2} values.
**Acknowledgments**
The present work was financially supported by a Junior Lecturer-Capacity Improvement grant (grant number 2458/UN1/DITLIT/DIT-LIT/PT/2021). The authors thank Dr. Titis Nurmasitoh, Ms. Hilliza Awalina, Mr. Suparno, Ms. Sumaryati and Ms. Dewi Sulistyowati for their valuable assistance during animal and specimen handling.
**Financial support and sponsorship**
This study was financially supported by the Junior Lecturer-Capacity Improvement Grant Universitas Gadjah Mada (grant number 2458/UN1/DITLIT/DIT-LIT/PT/2021).
**Conflicts of interest**
There are no conflicts of interest.
References | | |
1. | Xu J, Zhao S, Teng T, Abdalla AE, Zhu W, Xie L, *et al*. Systematic comparison of two animal-to-human transmitted human coronaviruses: SARS-CoV-2 and SARS-CoV. Viruses 2020;12:244. |
2. | Ji M, Su X, Liu J, Zhao Y, Li Z, Xu X, *et al*. Comparison of naturally aging and D-galactose induced aging model in beagle dogs. Exp Ther Med 2017;14:5881-8. |
3. | Stark A, Nyska A, Zuckerman A, Madar Z. Changes in intestinal tunica muscularis following dietary fiber feeding in rats. A morphometric study using image analysis. Dig Dis Sci 1995;40:960-6. |
4. | Casteleyn C, Rekecki A, Van der Aa A, Simoens P, Van den Broeck W. Surface area assessment of the murine intestinal tract as a prerequisite for oral dose translation from mouse to man. Lab Anim 2010;44:176-83. |
5. | Warren MA. Adaptations of the rat small intestine to a single and a double period of undernutrition. J Anat 1991;176:89-97. |
6. | Vdoviaková K, Petrovová E, Maloveská M, Krešáková L, Teleky J, Elias MZ, *et al*. Surgical anatomy of the gastrointestinal tract and its vasculature in the laboratory rat. Gastroenterol Res Pract 2016;2016:2632368. |
7. | Mayhew TM, Carson FL. Mechanisms of adaptation in rat small intestine: Regional differences in quantitative morphology during normal growth and experimental hypertrophy. J Anat 1989;164:189-200. |
8. | Boyce RW, Dorph-Petersen KA, Lyck L, Gundersen HJ. Design-based stereology: Introduction to basic concepts and practical approaches for estimation of cell number. Toxicol Pathol 2010;38:1011-25. |
9. | Mathieu O, Cruz-Orive LM, Hoppeler H, Weibel ER. Measuring error and sampling variation in stereology: Comparison of the efficiency of various methods for planar image analysis. J Microsc 1981;121:75-88. |
10. | Marcos R, Monteiro RA, Rocha E. The use of design-based stereology to evaluate volumes and numbers in the liver: A review with practical guidelines. J Anat 2012;220:303-17. |
11. | Ogiolda L, Wanke R, Rottmann O, Hermanns W, Wolf E. Intestinal dimensions of mice divergently selected for body weight. Anat Rec 1998;250:292-9. |
12. | Schroeder AB, Dobson ET, Rueden CT, Tomancak P, Jug F, Eliceiri KW. The ImageJ ecosystem: Open-source software for image visualization, processing, and analysis. Protein Sci 2021;30:234-49. |
13. | Bendtsen TF, Nyengaard JR, Grimelius L, Gundersen HJ. Stereological estimation of the total number of ECL cells and related parameters using the smooth, vertical fractionator in the rat oxyntic mucosa. J Microsc 2002;207:211-24. |
14. | Schmitz C, Hof PR. Recommendations for straightforward and rigorous methods of counting neurons based on a computer simulation approach. J Chem Neuroanat 2000;20:93-114. |
15. | Hassan ZA, Zauszkiewicz-Pawlak A, Abdelrahman SA, Algaidi S, Desouky M, Shalaby SM. Morphological alterations in the jejunal mucosa of aged rats and the possible protective role of green tea. Folia Histochem Cytobiol 2017;55:124-39. |
16. | Noorafshan A, Hoseini L, Karbalay-Doust S, Nadimi E. A simple stereological method for estimating the number and the volume of the pancreatic beta cells. JOP 2012;13:427-32. |
17. | Nyengaard JR, Alwasel SH. Practical stereology of the stomach and intestine. Ann Anat 2014;196:41-7. |
18. | de Brito MC, Chopard RP, Cury DP, Watanabe IS, Mendes CE, Castelucci P. Effects of aging on the architecture of the ileocecal junction in rats. World J Gastrointest Pharmacol Ther 2016;7:416-27. |
19. | Mohsenikia M, Rafiee S, Rozei LS, Ebrahimi A, Zahmatkesh-Meimandi F, Aref NM, *et al*. Althaea officinalis improves wound healing in rats: A stereological study. Drug Discov Ther 2020;14:239-42. |
20. | Gundersen HJ, Jensen EB, Kiêu K, Nielsen J. The efficiency of systematic sampling in stereology-reconsidered. J Microsc 1999;193:199-211. |
21. | Poorna Pillutla SV, Kaur C, Roy TS, Jacob TG. Estimation of volume of stria vascularis and the length of its capillaries in the human cochlea. J Microsc Ultrastruct 2019;7:117-23. |
22. | Mahmoudzadeh Sagheb HR, Moudi B. Basic Application of Stereology in Histology and Medical Sciences. Gene, Cell Tissue. 2014;1:e24237. doi: 10.17795/gct-24237. |
23. | West MJ. New stereological methods for counting neurons. Neurobiol Aging 1993;14:275-85. doi: 10.1016/0197-4580(93)90112-o. |
24. | Uribe A, Alam M, Johansson O, Midtvedt T, Theodorsson E. Microflora modulates endocrine cells in the gastrointestinal mucosa of the rat. Gastroenterology 1994;107:1259-69. doi: 10.1016/0016-5085(94)90526-6. |
25. | Hughes S, Lau J. A technique for fast and accurate measurement of hand volumes using Archimedes' principle. Australas Phys Eng Sci Med 2008;31:56-9. |
26. | Olesen MV, Needham EK, Pakkenberg B. The Optical Fractionator Technique to Estimate Cell Numbers in a Rat Model of Electroconvulsive Therapy. J Vis Exp 2017;(125):55737. doi: 10.3791/55737. |
**
**
**Correspondence Address**: Rina Susilowati, Department of Histology and Cell Biology, Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada, Sekip, Yogyakarta 55281 Indonesia
**Source of Support:** None, **Conflict of Interest:** None
**DOI**: 10.4103/jmau.jmau_53_22
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2] |