ABSTRACT:
OBJECTIVE Semiquantitative and automated measurement of nuclear material removal and cell infiltration
in decellularized tendon scaffolds (DTSs).
METHOD 16 pure New Zealand rabbits were used, and the gastrocnemius muscle tendon was collected
bilaterally from half of these animals (16 tendons collected); 4 were kept as control
and 12 were submitted to the decellularization protocol (DTS). Eight of the DTSs were
used as an in vivo implant in the experimental rotator cuff tear (RCT) model, and
the rest, as well as the controls, were used in the semiquantitative and automated
evaluation of nuclear material removal. The eight additional rabbits were used to
make the experimental model of RCT and subsequent evaluation of cellular infiltration
after 2 or 8 weeks, within the DTS.
RESULTS The semiquantitative and automated analysis used demonstrated a removal of 79% of
nuclear material (p< 0.001 and power > 99%) and a decrease of 88% (p < 0.001 and power >99%) in the area
occupied by nuclear material after the decellularization protocol. On cell infiltration
in DTS, an increase of 256% (p < 0.001 and power >99%) in the number of cells within
the DTS was observed in the comparison between 2 and 8 weeks postoperatively.
CONCLUSION The proposed semiquantitative and automated measurement method was able to objectively
measure the removal of nuclear material and cell infiltration in DTS.
Keywords: tendons; tissue engineering; tissue scaffolds; extracellular matrix; regenerative medicine.
RESUMO:
OBJETIVO Mensuração semiquantitativa e automatizada da remoção de material nuclear e da infiltração
celular em scaffolds tendinosos descelularizados (STDs).
MÉTODO Foram utilizados 16 coelhos Nova Zelândia puros, sendo o tendão do músculo gastrocnêmio
coletado bilateralmente de metade destes animais (16 tendões coletados); 4 foram mantidos
como controle e 12 foram submetidos ao protocolo de descelularização (STD). Dos STDs,
8 foram utilizados como implante in vivo no modelo experimental de lesão do manguito
rotador (LMR) e os restantes, assim como os controles, foram utilizados na avaliação
semiquantitativa e automatizada da remoção de material nuclear. Os oito coelhos adicionais
foram utilizados na confecção do modelo experimental de LMR e posterior avaliação
da infiltração celular após 2 ou 8 semanas, dentro do STD.
RESULTADOS A análise semiquantitativa e automatizada utilizada demonstrou uma remoção de 79%
do material nuclear (p< 0,001 e poder > 99%) e uma diminuição de 88% (p< 0,001 e poder > 99%) na área ocupada por material nuclear após o protocolo de descelularização.
Sobre a infiltração celular no STD, foi observado um aumento de 256% (p< 0,001 e poder > 99%) no número de células dentro do STD na comparação entre 2 e
8 semanas de pós-operatório.
CONCLUSÃO O método de mensuração semiquantitativo e automatizado proposto foi capaz de mensurar
objetivamente a remoção de material nuclear e a infiltração celular no STD.
Palavras-chave: tendões; engenharia tecidual; tecidos suporte; matriz extracelular; medicina regenerativa.
FIGURES
| Citation: Santos AL, Silva CG, Barreto LSS, Tamaoki MJS, Almeida FG, Faloppa F. Automated Assessment of Cell Infiltration and Removal in Decellularized Scaffolds – Experimental Study in Rabbits. 57(06):992. doi:10.1055/s-0041-1739174 |
| Financing Support The present study was funded by the National Council for Scientific and Technological Development (CNPq) – process number 311237/2018-5. |
|
Conflict of Interests: The authors have no conflict of interests to declare. |
|
Study developed at Escola Paulista de Medicina da Universidade Federal de São Paulo, São Paulo, SP, Brazil |
| Received: October 30 2020; Accepted: June 25 2021 |
INTRODUCTION
Rotator Cuff Tear (RCT) is a frequent cause of shoulder pain and may present clinically with a wide diversity of symptoms, with pain, weakness, and movement limitation being the most frequent.1 Regarding its treatment, there are currently conservative or surgical options,2 and their choice is performed according to the following criteria: (i) characteristics of the patient, (ii) morphology of the lesion, and (iii) main complaints.3 In the morphological evaluation, massive lesions are related to inferior postoperative outcomes than small.4
Seeking to optimize the results of surgical treatment of RCT with great retraction, there is a strong tendency to use tendon substitutes, whether decellularized or not.5 Conceptually, the tendon substitute is a tissue similar to the tendon and should perform its function when implanted in vivo.6 To be considered ideal, it must present the following characteristics: (a) structure in three dimensions with high porosity; (b) minimal cellular material, seeking to avoid inflammatory response; (c) cytocompatibility; and (d) biomechanical properties suitable to support the mechanical needs of rehabilitation up to complete cell infiltration and healing.7 8
Tendon substitutes are commonly known as grafts; however, recently, the concept of scaffold has been widespread. Scaffolds are tissues that must allow cell infiltration and replication within their structure9 and can be used as grafts when applied in vivo. They can be classified, according to their origin and composition, as synthetic or biological. Synthetics are industrialized polymers and present results with great variability according to the chosen material, function, and technique performed.10 11
Regarding biological substitutes, they are subdivided according to their origin into autogenous, allogenous and xenogenous.7 12 Autogenous grafts are the gold standard; however, their low availability is the main obstacle to their use.7 Xenogenous and allogenous grafts, on the other hand, present, as adversity, uncontrolled inflammatory response and unsatisfactory tissue integration. Seeking to optimize the inflammatory response, to facilitate tissue integration, and to maximize the availability of biological substitutes, there is a strong tendency for its processing through decellularization.13 14 Decellularization is a processing that must combine physical, chemical, and enzymatic techniques and has the ability to remove cellular material.15 16 17
A decellularized tendinous scaffold (DTS) was recently evaluated in a Brazilian experimental study and presented maintenance of the main biomechanical properties and substantial removal of nuclear material, besides allowing cell infiltration.18 Regarding the removal of nuclear material and cell infiltration, the literature usually recommends descriptive analyses,13 15 19 20 and quantitative evaluation21 is one of the recent possibilities to optimize the validity of this evaluation.
Thus, in the search for alternatives to measure cell infiltration or removal, the present study proposes to evaluate the hypothesis that the semiquantitative and automated methodology presented can perform the measurements proposed in DTSs.
METHOD
Experimental Design
Sixteen male rabbits (Pure New Zealand Rabbit, Granja RG-PR, Suzano, SP, Brazil) weighing between 3 and 3.5kg, maintained at the Center for the Development of Experimental Models for Medicine and Biology (CEDEME, in the Portuguese acronym) were use. The animals used were part of the validation of the previously published decellularization process,18 and the present study refers to the presentation of a new methodology for semiquantitative and automated measurement of nuclear material removal and cell infiltration of these same specimens. During the experiments, the animals remained in individual cages, with a light/dark cycle of 12/12hrs, food and water ad libitum. 22 The study was approved by the Ethics Committee on the Use of Animals (CEUA527208916), followed the guidelines proposed by the ARRIVE guideline,23 and received support from the National Council for Scientific and Technological Development (CNPq, in the Portuguese acronym) (311237/2018–5).
To perform the experiments, the animals were divided into two different designs: in the first, the gastrocnemius muscle tendon was collected, and the DTS was prepared. A total of 16 tendons of the gastrocnemius muscle (corresponding to 8 animals) were collected, 12 of which were submitted to the decellularization protocol (DTS), and the other 4 were maintained as controls. Eight DTSs were inserted in the experimental model in vivo and four were used in the histological evaluation (►Figure 1). The remaining eight animals (►Figure 1) were used in the experimental models of RCT, as shown below.
Preparation of Scaffold from Decellularized Tendon
The decellularization protocol used for DTS production, which has been published and validated previously,18 presents the following steps: the gastrocnemius muscle tendons were washed with phosphate buffered saline (PBS) solution containing 1% antibiotic (penicillin-streptomycin solution; Sigma-Aldrich, St. Louis, MO, USA) for removal of surface residues and of the decellularizing agent used in the previous stage.
The remainder of the protocol included maintenance of the specimens in constant agitation (MaxQ4000; Thermo Scientific, Waltham, MA, USA) and subsequent exchanges of the following decellularizing agents: aprotinin (Sigma-Aldrich St. Louis, MO, USA), ethylenediaminetetraacetic Acid (EDTA) (Sigma-Aldrich, St. Louis, MO, USA), sodium sulfate (SDS) (Invitrogen, Carlsbad, CA, USA), and t-octil-phenoxypolyethoxyethanol (Triton X-100) (Affymetrix, Maumme, Ohio, USA).
In vivo Experimental Model of Rotator Cuff Injury
For the preparation of the experimental RCT model, the animal was submitted to the following anesthesia and analgesia protocol: initial analgesia and preoperative antibiotic therapy with tramadol (5 mg/kg) and terramycin (50 mg/kg); after 30 minutes, anesthesia was started with ketamine (50 mg/kg) and xylazine (10 mg/kg). In the postoperative analgesia, the animal was maintained with meloxican (0.5 mg/kg) and tramadol (5 mg/kg) until completing the 3rd postoperative day, and these medications were administered in case of pain or discomfort after this period. Evaluations regarding stress, discomfort, and pain were performed daily at CEDEME.
After anesthesia, the animals were positioned in the supine position and submitted to bilateral glenohumeral joint trichotomy, asepsis, and antisepsis. Thus, both front paws were submitted to the RCT protocol and then allocated into the “Injury” or “Injury + DTS” groups. For allocation, simple randomization was performed, so that one of the paws was used as the contralateral control.
For the injury protocol, an anterolateral pathway was performed in the shoulder, with exposure and dissection of the deltoid muscle between its anterior and middle sections (►Figure 1a). After dissection of the interval and exposure of the subscapularis tendon (►Figure 2b), a lesion parallel to the tendon fibers was performed throughout their extension (►Figure 2c), without disinsertion in the ostenotendinous junction24 (►Figure 2d). After the injury, the DTS was positioned in one of the paws, exactly on the site of the experimental lesion, and had its extremities fixed with nylon (Nylon 4–0; Shalon, Alto da Boa Vista, GO, Brazil). The contralateral was submitted only to the marking of the lesion. In this experimental model, we limit the biomechanical requirements to which the DTS would be subjected, thus ensuring that there would be no failures in the fixation of the DTS to the rotator cuff. However, the results obtained would be limited to the inflammatory response and tissue integration between the DTS and the rotator cuff.
The animals were again randomized in relation to postoperative time (2 or 8 weeks postoperatively) and submitted to painless induced death, with an overdose of anesthetics (ketamine 200mg/kg + xylazine 40 mg/kg and tramadol 10 mg/kg). Then, the previously marked region was collected, necessarily including the rotator cuff throughout the lesion and the DTS.
Preparation and Coloring with Hematoxylin and Eosin
The central part of the DTS (n =4), of the control (n =4), and the material resulting from the in vivo analysis (Injury [n =8] and Lesion + Scaffold [n =8]) were prepared through the following protocol: fixation in 10% formaldehyde, dehydrated with ethyl alcohol, diaphanized by xylol, and impregnated with liquid paraffin. Manual inclusion and positioning of the microtome blocks was then performed for cuts with a thickness of 4 μm and a distance of 50 μm. Prior to staining with hematoxylin and eosin (H&E), the sections were deparalinated with xylol, hydrated with ethyl alcohol concentrations and immersed in distilled water. For staining with the H&E technique, immersion in hematoxylin solution, the sections were washed in running water and dehydrated with ethyl alcohol until they were colored with eosin. The H&E-colored histological slides were evaluated using the Olympus IX 81 optical microscope (Olympus Corporation, Shinjuku-ku, Tokyo, Japan) with fluorescence, while the images were captured by an Olympus DP72 camera (Olympus Corporation, Shinjuku-ku, Tokyo, Japan) coupled to the microscope.
Semiquantitative and Automated Analysis of Nuclear Material Removal and Cellular Infiltration
The analysis of nuclear material, be it removal or cell invasion, was performed using a semiquantitative methodology,21 with an automated method. In this methodology, the removal of nuclear material was classified as: complete (100%), substantial (99–70%), moderate (69–50%), minimum (49–30%) and no change (< 30%).6 For the semiquantitative and automated analysis of nuclear material removal, the slides referring to four DTS and four controls were performed and in a way of evaluate cell infiltration, the same method was used to quantified the number of nucleus in four Lesions + DTS with 2 weeks and another four Lesion + DTS with 8 weeks postoperatively.
To measure cell infiltration, we applied the methodology only in the region corresponding to the DTS. For counting, 10 random fields were photographed clockwise from superior to lower with an increase of 400x, thus totaling 40 control photographs, 40 of DTS, 40 of the Lesion + DTS group with 2 weeks, and 40 of the Lesion + DTS group with 8 weeks postoperatively.
The photographs were inserted in Image J software (Image J 1.53e; National Institutes of Health, Bethesda MA, USA), and the scale was adjusted in the first image evaluated. To start the counting, the color was decombed using the “colour deconvolution - H&E2” plugin (►Figure 3) and only the image with eosin color enhancement (►Figures 3B and ►Figure 4C) was selected, which highlights the nuclear material. The figures were then manually adjusted in the topic image - adjust - threshold and then converted to the binary format (process - binary - convert to mask -make binary) (►Figure 4d). As a final step, automated counting was performed using the analyze function – analyze particles (size 1–100μm2; circularity 0.00–1.00; outlines; exclude on edges; include holes) (►Figure 4b). The software provided the following results at the end of the process: total nucleus count, total nucleus area, average area of the nucleus, and percentage of area of each photograph occupied by the nucleus.
Statistical analysis
Using the available literature,21 we found a similar decellularization protocol, with removal of 97.5% (±2%) of nuclear material. Applying this difference between the groups, through the two-sample t test for mean differences, considering an alpha of 0.05 and a power of 95%, the minimum number of specimens per group was 2. For statistical difference evaluation, the Wilcoxon-Mann-Whitney nonparametric test was chosen using SAS Studio software (SAS 3.8 Basic Edition; SAS Institute, Cary, NC, USA), and p <0.05 is presumed as a statistically significant difference. As a mechanism to evaluate the results obtained, we also chose to evaluate in a post hoc way (satterthwaite t-test) the statistical power of the results found, using SAS Studio software (SAS 3.8 Basic Edition; SAS Institute, Cary, NC, USA). Microsoft Excel for Office 365 (Microsoft Corporation, Redmond, WA, USA) was used to assemble the tables (descriptive presentation) and prepare the graphs.
RESULTS
Measurement of Decellularization by Semiquantitative and Automated Method
A substantial decrease in nuclear material was observed (►Figure 6), since ∼ 79% (p< 0.0001 and statistical power > 99%) of this material was removed during decellularization (►Figure 6 and ►Table 1). Also in the semiquantitative analysis, we noticed a reduction of ∼ 88% (p< 0.001 and statistical power > 99%) in the area occupied by nuclear structures after the decellularization protocol, confirming the substantial removal of nuclear material (►Table 2).
| Nucleus count (SD) | p-value | |
|---|---|---|
| Control | 248.85 (91.53) | < 0.0001 |
| DTS | 50.45 (24.35) | |
| RCT + DTS - 2 weeks PO | 57.90 (50.49) | < 0.0001 |
| RCT + DTS - 8 weeks PO | 148.67 (82.29) |
Difference between control and DTS demonstrates substantial removal of nuclear material with statistical significance. Similarly, the increase in the number of nuclei over the postoperative period demonstrates progressive cellular infiltration in the DTS, with statistical significance.
| Nucleus count (SD) | p-value | |
|---|---|---|
| Control | 2.628% (1.094%) | < 0.0001 |
| DTS | 0.293% (0.175%) | |
| RCT + DTS - 2 weeks PO | 0.558% (0.552%) | < 0.0001 |
| RCT + DTS - 8 weeks PO | 1.795% (1.353%) |
Difference in the area occupied by nuclear material between control and DTS demonstrates substantial removal of nuclear material with statistical significance. Similarly, the increase of this area over the postoperative period demonstrates progressive cellular infiltration in the DTS, with statistical significance.
Measurement of Cellular Infiltration in DTS After Implant In Vivo
The DTSs were easily characterized, in macroscopic and histological vision, in the different postoperative moments (►Figures 7 and 8). Throughout the postoperative period, the DTS was in progressive integration with the rotator cuff, since it was possible to notice cellular infiltration still restricted to the peripheries of the DTS after 2 weeks, and a more comprehensive infiltration after 8 weeks postoperatively (►Figure 8).
In the automated measurement, the DTS presented an average of 50.45 cells, a number that remained stable after implantation in vivo and after 2 weeks postoperatively (p= 0.6081) and increased significantly to 148.67 cells (p< 0.001 and power > 99%) after 8 weeks postoperatively (►Table 1 and ►Figure 9).
In the measurement of the area, there was a similar increase: the DTS presented 0.175% of the area occupied by nuclei; after 2 weeks, this area remained stable (p= 0.1611), and it increased to 1.795% (p < 0.001 and statistical power > 99%) 8 weeks postoperatively (►Table 2).
DISCUSSION
The main findings of the present study are the presentation of an innovative, semiquantitative and automated method for measuring the removal of nuclear material and cellular infiltration in decellularized tissues produced in Brazil and that are possible to be evaluated for tendon replacement in the most diverse lesions, possibly in RCT with great retraction.
The search for tendon substitutes is not recent, since the problem related to RCT with great retraction is ancient. Currently, the use of autologous material is presented as the gold standard;25 26 however, the limitations related to its use (biomechanical inability to replace the original tissue) and collection (morbidity at the donor site and limited availability) motivate the search for new alternatives.
Considering these limitations, it is essential to search for alternative tissues that can perform functions similar to those of the tendon, and decellularization is the most promising method.16 27 28 In this sense, the measurement of the reduction of nuclear material in the processing and favoring of cellular infiltration after in vivo implantation are important objectives to be used in subsequent studies.
These studies should seek to develop an ideal tendon substitute, overcoming the current limitations of biologicals; thus, it should be a product with acceptable biocompatibility, no risk for disease transmission, low risk for chronic inflammatory response, and storage capacity for long periods.
The main limitations of the present study are related to the small number of specimens evaluated, the absence of additional evaluations, such as the objective measurement of DNA and remaining collagen, and the characteristics of the experimental model. The experimental lesion used does not reproduce the conventional lesions of the rotator cuff; however, it favors tendon healing and tissue integration in the DTS/tendon interface.
As a strong point of the present study, the use of DTS is promising in Brazil, since Brazilian legislation does not allow the commercialization of some tissues with an allogenous origin used in other countries. In this perspective, the standardization of decellularization in a center for human tissue management may provide a viable alternative for future studies of patients with RCT.
CONCLUSION
The proposed semiquantitative and automated measurement method was able to objectively measure the removal of nuclear material and cellular infiltration in the scaffold.
