Study on the changes of ADRP+ cell population in the process of pulmonary brosis

ADRP is a marker of lung lipobroblasts. Lipobroblasts play an important role in assist type 2 alveolar epithelial cells function in peripheral lung tissue. Pulmonary brosis is characterized by continuous irreversible destruction of peripheral lung tissue. The expression of ADRP and the role of ADRP + cells in pulmonary brosis are of interest to us. Quantitative PCR as well as immunohistochemical multiplex staining were used to analyze the expression of ADRP during lung development, bleomycin-induced pulmonary brosis, and identify the type and function of ADRP + cells during pulmonary brosis lesions.


Abstract
Background ADRP is a marker of lung lipo broblasts. Lipo broblasts play an important role in assist type 2 alveolar epithelial cells function in peripheral lung tissue. Pulmonary brosis is characterized by continuous irreversible destruction of peripheral lung tissue. The expression of ADRP and the role of ADRP + cells in pulmonary brosis are of interest to us.

Methods
Quantitative PCR as well as immunohistochemical multiplex staining were used to analyze the expression of ADRP during lung development, bleomycin-induced pulmonary brosis, and identify the type and function of ADRP + cells during pulmonary brosis lesions.

Results
ADRP + cells were found to decrease gradually from birth to adulthood. During pulmonary brosis, the expression of ADRP increased gradually, while another marker of lipo broblast, the expression of PDGFRα decreased. There was a co-localization relationship between macrophage marker CD68 and ADRP. Both M1-type and M2-type macrophages can express ADRP in the early stage of pulmonary brosis. While in the subsequent pulmonary brosis process, M1-type ADRP + macrophages gradually decrease, while the ADRP + cells were mainly M2-type macrophages. ADRP + M2-type macrophages can release S100A4 and distribute around the lesion area of pulmonary brosis. Pulmonary brosis gradually developed as M2-polarized macrophages replaced M1 to become the main population of pulmonary ADRP + macrophages

Conclusion
In the process of pulmonary brosis, a large number of ADRP + cells are macrophages, including M1 and M2 types successively. The aggregation site of ADRP + M2-type macrophages indicates that brosis damage is about to or has already occurred. This study may provide a new research direction for the treatment of pulmonary brosis.

Background
Adipose differentiation related protein (ADRP) belongs to Perilipin protein family, is also called Perilipin 2 or Plin2. ADRP distributed on the surface of neutral lipid droplets in cytoplasm. Cytoplasmic lipid droplets degradation can be prevented when ADRP encapsulated. This results in accumulation of intracytoplasmic lipid droplets. ADRP plays an important role in the balance between lipid droplets storage and lipid out ow [1]. On the other hand, as a marker of cell differentiation, ADRP expression is strongly induced to form ADRP + cell when cells are faced with increased lipid load [2]. For example, mononuclear macrophages engulf a large amount of lipids to form foam cells in the atherosclerotic plaque, and ADRP is the marker protein of foam cells [3]. Knocking out ADRP gene affects the formation of foam cells and inhibits atherosclerosis [4]. Changes in ADRP expression patterns may indicate changes in the course of brosis. Decreased ADRP expression is associated with lipid loss and brosis in hepatic stellate cells [5]. Inducing the expression of ADRP can inhibit the expression of genes related to liver brosis in cultured cells, while inhibition of ADRP expression can reverse this process. [6] ADRP is also expressed in lung tissue. There are ADRP + cells in the lung interstitial tissue of normal newborn mice. These cells called lipo broblasts for their cytoplasm contain neutral lipid droplets. Distributed near type II alveolar epithelial cells, lipo broblasts play important role in lipid uptake and transport to type II alveolar epithelial cells, affecting the synthesis and secretion of surfactant [7,8].
Lipo broblasts contain active substances such as retinoic acid, which are closely related to alveolar growth, differentiation, homeostasis and repair. Decreased lipo broblasts in newborn mice may lead to bronchopulmonary dysplasia (BPD) [9]. In the interstitial tissue of the terminal lung, the composition of lipo broblasts is complex. Lipo broblasts express many types of marker molecules, such as PDGFRα, CD90, PPARγ, and ADRP [10,11]. Among these molecular markers, ADRP is often used as a target for immunohistochemical examination of mature lipo broblasts due to its coexistence with neutral lipid droplets. Most often, lipo broblasts are ADRP + cells in normal lung tissues.
Pulmonary brosis is a progressive and irreversible destruction of alveolar structures in peripheral lung tissues. In this process, the role of lipo broblasts is of concern. It has been reported that lipo broblasts transdifferentiated into myo broblasts during pulmonary brosis induced by one-time perfusion of bleomycin through the airways, while lipo broblasts reappear when brosis is alleviated [12]. Based on scRNA-seq technique, recent study has shown that pulmonary lipo broblasts can also express immune function-related genes and some M2 macrophage-related genes [13]. Unlike one-time Bleomycin administration in the airway, which induces self-healing pulmonary brosis, multiple intraperitoneal administration causes repeated lung tissue damage and repair, which triggers irreversible brosis which is more similar to clinical pulmonary brosis to some extent [14]. In this case, we are interested in the expression pattern of ADRP, the fate of ADRP + cells, and the relationship between ADRP + cells and brosis. We found that macrophages can form intrapulmonary "ADRP + cells", similar to foam cells in atherosclerotic plaques, and affect lesions in adjacent lung tissues.

Methods
Animal grouping, pulmonary brosis model preparation and lung section randomly divided into 5 groups with 6 mice in each group, including 1 control group and 4 experimental groups which were induce pulmonary brosis. Experimental mice were intraperitoneally injected with Bleomycin according to the following protocol that is 1 USP/mouse on days 0 and 2, and 0.5 USP/mouse on days 4 and 6, and 0.25 USP/mouse on days 9, 12, 15, 18, 21, 24, and 27. The experimental groups were harvested on the 7th, 14th, 21st and 28th day respectively to observe the course of disease. The control group was injected with the same volume of normal saline and collected on the 28th day.
To evaluate the effect of different number of injection times on induced pulmonary brosis in mice, we truncated modeling process. The mice were randomly divided into 6 groups with 6 mice in each group, including 1 control group and 5 experimental groups. According to the modeling protocol, the mice in the ve experimental groups completed 1,2,3,4 and all 11 times intraperitoneal Bleomycin injection respectively. The normal control group was injected with the same volume of normal saline. All six groups mice lung were collected on the 28th day.
Mice were sacri ced by cervical dislocation and the dissected. Normal saline is injected from the right ventricle for pulmonary perfusion to clear the blood from the lungs until they become white. Lung was lled with 4% paraformaldehyde of 20 cm of water pressure, then it was ligated and xed in a container containing the 4% paraformaldehyde for 16 hours at 4℃. Alcohol gradient dehydration, xylene transparent, and para n-embedded. Cut 5 µm thick tissue sections HE staining, Masson staining HE staining: Lung tissue sections were depara nized and rehydrated, stained with hematoxylin for 1 min, washed with distilled water for 20-30 times, differentiated with 0.5% ethanol for 20 s, washed with distilled water for 20 s; blue with 1% aqueous ammonia for 10 s, washed with distilled water for 20-30 times; washed with 95% ethanol two times, and stained with eosin for 1 min. Then the tissue sections were dehydrated conventionally, transparent, and sealed with neutral gum, then observed under the microscope.
Masson staining: The lung tissue sections were depara nized and rehydrated, washed with water; stained with orcein staining solution for 20 min, washed with water; stained with iron hematoxylin staining solution for 10 min; differentiated with hydrochloric acid ethanol for 20 s, washed with water; returned to blue with 1 × PBS for 20 s, washed with water; stained with Ponceau S -acid fuchsin staining solution for 10 min, washed with 0.2% glacial acetic acid for 1 min; color separation with 0.5% phosphomolybdic acid for 40-45 s, washed with 0.2% glacial acetic acid for 1 min; stained with 0.5% aniline blue staining solution for 1 min, washed with 0.2% glacial acetic acid for 1 min; blotting up excess staining solutions with lter paper; the tissue sections were routinely dehydrated, transparent, sealed with neutral gum, and dried, then observed under the microscope.
Random observation of lung tissue visual eld of three different lung brosis mice under 10 × 20 times microscope. If the cell is not stained, score 0, if the cell is stained, score 1. Taking the average value of the integral arithmetic of 3 elds as the nal integral. The ADRP + and CD68 + cells and co localization cells were counted, and the proportion of ADRP + CD68 + /ADRP + cells in different periods was obtained. The same method was used to obtain the ratio of ADRP + CD86 + /ADRP + cells and ADRP + CD206 + / ADRP + cells in different periods.
Immunohistochemical antibody stripping multiple staining Photographs were taken under microscope after the rst immunohistochemical staining. Wash the AEC stain with gradient alcohol (25%-95%). Antibodies were stripped in the buffer containing 65 mM Tris-HCl pH6.8(Sigma, USA) 1% SDS Solarbio, USA 0.113 M 2-mercaptoethanol Sigma, USA 0.1 M NaCl 2 M Urea Sigma, USA ,in 56℃ water bath with agitation in the fume hood twice for 30 minutes each. After antibody stripping, the sections were washed in distilled water for 0.5 hours and replaced every 5 minutes in room temperature. Then the sections can be used for immunohistochemistry staining again according to the above protocol. Without adding rst antibody, a negative control was setup to avoid any false staining due to incompletely antibody stripping.

Frozen tissue sections, Oil red O staining and Immunochemistry Stain
The model mice were sacri ced and the lungs were collected, perfused, and xed with 4% paraformaldehyde for 16 hours following the above protocol. Then the lung samples were immersed in sucrose solutions with concentration gradients of 10%, 20% and 30%, respectively, for 24 hours in each gradient. Treated lungs were embedded with OTC and stored in -80 °C refrigerator, and Cut into 10 microns thick tissue sections under − 25℃.
The lung tissues frozen sections were stained with diluted Oil red O staining solution (Solarbio, USA) for about 10 minutes, 60% isopropanol color separation to the background colorless, counterstained with hematoxylin for 1 minute, washed with 1 × PBS, sealed with 70% glycerol and then took photograph. Next, the sections washed with 60% isopropanol and then ethanol hydrochloride washing to remove any color. The sections were used to stain ADRP following above immunohistochemistry protocol.

Realtime PCR
RNA was extracted from the left lobe of mouse lung. RNA samples were reverse transcribed into cDNA and then ampli ed and quantitated with Realtime PCR. The program was run on the ABI7500 Realtime PCR System instrument using TaKaRa SYBR PremixExTaq reagent (Cat# RR820A). All experiments were repeated 3 times. The measurement data were expressed as mean±standard deviation (mean±SD). The composition analysis was performed using GraphPad Prism5 data analysis software. Comparison between multiple groups was performed by single factor analysis of variance.

ADRP + cell in mice lung development and pulmonary brosis
ADRP is generally considered to be a marker molecule of lipo broblasts in the lung. We examined the distribution of ADRP + cells in the lungs of mice at different developmental stages by immunohistochemistry. ADRP + cells gradually decreased with the formation of alveoli after birth and the ADRP + cells were very rare in normal adult mice lung (Fig. 1a). Pulmonary brosis was induced in adult mice by intraperitoneal injection of bleomycin (FigS1). Lung samples were taken at the 7th (D7), 14th (D14), 21st (D21) and 28th days (D28) after modeling. The expression levels of some important cell differentiation markers were detected by Realtime-PCR. Compared with control group, α-SMA and Collagen gene expression levels increased signi cantly from the beginning of the D21. This was associated with increased myo broblasts leading to pulmonary brosis. Lipo broblasts related marker gene PDGFRα expression level decreased signi cantly from the 7th day, until D28. However, in contrast to PDGFRα, the expression of ADRP in D28 days was signi cantly increased compared with that in the control group (Fig. 1b).
We further examined the ADRP in lung tissue on the 7th and 28th day after modeling by immunohistochemical staining. Compared with the sample on the 7th day the number of ADRP + cells on the 28th day' sample was increased (Fig. 1c). Lung tissue frozen sections were prepared from the 28th day's sample and stained with Oil red O. Oil red O positive stained cells were found. After Oil red O staining was removed from the same section, immunohistochemical staining showed that the positive cells with Oil red O staining were ADRP + cells (Fig. 2). Found in the lung tissues of brosis model mice and containing neutral lipid particles, these ADRP + cells have different distribution and shape from the general understanding lipid broblasts. Many of them were located in the alveolar cavity, with large diameter and single nucleus. These cells have morphological characteristics similar to that of macrophages in alveolar cavity.

During pulmonary brosis ADRP + macrophage polarization pattern changes
We analyze the expression level of macrophage-associated molecular markers CD68 CD86 CD206 on the 7th ,14th ,21st and 28th days during the process of pulmonary tissue lesion in brosis model mice with Realtime-PCR techniques. It was found that the expression level of macrophage molecule marker CD68 gradually increased during the modeling process, and the signi cant difference began to appear on the 14th day after the modeling compared with that of the control group. The expression of M1-polarized macrophage molecular marker CD86 was also increased, and the signi cant difference began to appear on the 28th day after modeling. Although the expression of M2-polarized macrophage molecular marker CD206 was increased in the same period, but the difference was not signi cant (Fig. 3a).
To identify whether ADRP + cells are macrophages, we examined the co-localization relationship of ADRP with CD68 CD86 and CD206 on the 7th ,14th ,21st and 28th days of modeling mice lung section using multiple immunohistochemical staining. It was found that the ADRP + cells increased gradually in the lung tissues of the model mice. These cells distributed in the lung tissues and alveolar cavity. In total ADRP + cells, the proportion of ADRP + CD68 + cells were close to 80% on the 7th day and gradually increased to almost 100% on day 28 (Fig. 3b). There is no signi cant difference of ADRP + CD68 + cells proportion in these two time points. About half of the ADRP + cells were ADRP + CD86 + cells on the 7th day in modeling mice lung section, and then to D28, the ADRP + CD86 + cells gradually approached to almost disappear ( Fig. 3C). In the lung sections of mice collected on the 7th day after modeling, about half of the ADRP + cells were ADRP + CD206 + , while most of ADRP + cells were ADRP + CD206 + on day 14 and it lasted until the 28th day (Fig. 3d).
In summary, there is an ADRP + macrophage polarization pattern changes during pulmonary brosis. In the early stage of bleomycin induced lung injury (D7), most of the ADRP + cells were macrophages, that is M1-type or M2-type. As the damage continued (D14), the proportion of ADRP + M2 macrophages exceeded that of ADRP + M1 macrophages and nearly all ADRP + cells were M2 macrophages after the peak of brosis (D21 and D28) in the model mice.

M2 polarized ADRP + macrophage associated with pulmonary brosis
We detected ADRP, N-cadherin and α-SMA expression pattern in serial sections of the 21st day of modeling mice lung tissue by immunohistochemical staining. There were sporadic brosis lesions in the lungs of model mice. A large number of α-SMA and N-cadherin positive cells were found in the brosis lesion area. ADRP + cells distributed in the area adjacent to brotic foci (Fig. 4a). While in the area without brosis lesion, no ADRP + cells were found (Fig. 4b). This indicating that the appearance of M2 polarized ADRP + macrophages is associated with brotic lesions.
In order to analysis of the relationship between ADRP + macrophage appearance and the degree of lung injury, we reduced the number of intraperitoneal bleomycin administration, and only completed the rst, second, third and fourth intraperitoneal injections respectively. After completion of the corresponding injection, modeled mice were fed to the 28th day and then lung samples were collected. Fibrosis foci were found in lung histopathology specimens of modeled mice which received three or more times intraperitoneal bleomycin injection (FigS2). There are aggregated α-SMA + cells in brosis foci. ADRP + cells distributed around the brosis foci (Fig. 5a). However, the lung structure was normal in the model mice which were injected bleomycin only 1 or 2 times, and there were no brotic foci and ADRP + cell aggregation (FigS2).
The expression level of CD68 gene, a macrophage marker, increased with the number of injections. The expression of lipo broblast marker PDGFRα showed a signi cant decreasing trend after 2 administration, while the expression of ADRP increased signi cantly after 4 times administration (Fig. 5b). We examined the co-localization relationship of ADRP with CD206 in modeling mice with three intraperitoneal injections using multiple immunohistochemical staining. The results showed that ADRP + M2 macrophages appeared around the brosis lesions (Fig. 5c). It is suggested that the formation of ADRP + M2 macrophages is related to pulmonary brosis again.
4. M2-polarized ADRP + macrophages were present at the site of peripheral lung tissue injury The proliferation and differentiation status of ADRP + cells were analyzed using multiple immunohistochemical staining. We found that PCNA was not expressed in ADRP + cells in D21 model mice lung tissue. This indicate that ADRP + cells did not proliferate. But adjacent to these ADRP + cells, some of the interstitial cells expressed PCNA and showed proliferative status (Fig. 6a). We noticed that most of the ADRP + cells expressed Caspase9. Moreover, in adjacent regions of ADRP + cells, many alveolar epithelial cells were also expressed Caspase9, suggesting that this was the alveolar epithelial injury region (Fig. 6b). Furthermore, a large number of ADRP + S100A4 + double positive cells were observed in the lung tissue of D21 model mice (Fig. 6c). It suggests that M2-polarized ADRP + macrophages may affect adjacent cells by releasing S100A4.

Discussion
Our work con rms that ADRP cannot be used as a marker for lipo broblasts in studies of pulmonary brosis. Pulmonary brosis leads to irreversible destruction of pulmonary alveolar structure.
Lipo broblasts, a kind of lung interstitial cells, are generally considered to play an important role in the formation and stabilization of alveolar structure. Therefore, appropriate molecular markers of lipo broblasts should be selected for the study of pulmonary brosis. A variety of lung interstitial cells express PDGFRα. Some of these PDGFRα + cells are involved in the cell niches that maintains AEC2 in appropriate differentiation status [15,16]. And CD34 + PDGFRα + lung interstitial cells contain high levels of neutral lipids [17]. Thus, lipo broblasts are included in the pulmonary PDGFRα + cell population. We found that the expression level of PDGFRα was signi cantly lower than that of control group in model mouse lung. This indicated that PDGFRα + cells, including lipo broblasts, decreased or disappeared due to transdifferentiation or apoptosis during pulmonary brosis process. It has been reported that PDGFRα + cell lineage involved in the formation of brotic lesion [18]. The decrease or even disappearance of PDGFRα + cells directly altered the cell niches required for the normal functioning of AEC2 and affected the homeostasis of the terminal lung tissue. A very contradictory result is that the classical lipo broblast marker molecule ADRP expression is signi cantly elevated in model mouse lung compared with that of control group. The reason for this phenomenon is that ADRP is a differentiation marker related to cell lipid load, and a group of high lipid load ADRP + cells appeared in the lungs of model mice. Our experiments con rmed that a large amount of lipids was stored in cytoplasm of ADRP + cells (Fig. 2), and ADRP + cells express macrophages marker (Fig. 3). Many ADRP + cells were distributed in the alveolar cavity, with typical morphology of macrophages. It can be inferred that intraperitoneal injection of bleomycin caused damage to the peripheral lung tissues. Macrophages in the lung tissues engulfed lipidrich alveolar surfactant and apoptotic alveolar epithelial cells. The lipid load of macrophages increased.
This results in signi cantly increased expression of ADRP, eventually forming ADRP + cells in the lung.
This process may be similar to the formation of ADRP + foam cells during atherosclerosis [19]. Study based on single-cell sequencing has found that lipo broblasts cells can express the immune-related gene CD206 of M2-type macrophages [20]. Although the modeling method for inducing brosis is different from ours, it cannot rule out that a considerable part of ADRP + cells are macrophages rather than intended lipo broblasts.
Our work con rms that the ADRP + macrophages polarization mode is associated with the course of pulmonary brosis. Macrophages are an important part of the body's immune defense system. In the modeling process of pulmonary brosis induced by intraperitoneal injection of bleomycin, ADRP + macrophages changed regularly with the course of the disease. Intraperitoneal injection of bleomycin rst caused an in ammatory response in the lungs, followed by brosis and progressive aggravation [14]. We Pulmonary brosis lesions, formed after a one-time intratracheal perfusion of bleomycin administration in mice, were self-healing over time delay. While multiple intraperitoneal injections of Bleomycin caused repeated tissue damage and repair, it will trigger irreversible brosis. The course of pulmonary lesions is similar to that of brosis observed clinically [14]. To verify the association of ADRP + M2 macrophage aggregation emergence with pulmonary brosis, we truncated the modeling process. It was found that 1-2 times intraperitoneal injections of Bleomycin did not cause signi cant and persistent pulmonary brosis injury, and no ADRP + cells aggregation. At least 3 intraperitoneal injections of bleomycin triggered irreversible pulmonary brosis injury. Signi cant expression of α-SMA was observed in lung tissues (Fig. 5). There is ADRP + cells aggregation in modeled mice lung (Fig. 5). Again, ADRP + M2 macrophages were associated with pulmonary brosis.
We note that many ADRP + macrophages express Caspase9 which suggests that these cells may undergo apoptosis in the future (Fig. 6). This also explains that although pulmonary brosis lesions can be detected in D28 lung tissue sections after three injections, the number of ADRP + macrophages in the lung was small. More injections of bleomycin are needed to induce more extensive brosis damage in order to see a signi cant increase in ADRP by Realtime-PCR in the total RNA extracted from lung tissue (Fig. 5).
We found that brosis damage is imminent or has occurred at ADRP + macrophage aggregation sites.
Lung specimens of model mice showed that some ADRP + cells and their adjacent epithelial cells expressed Caspase9 (Fig. 6). This is a signal of cell injury and apoptosis. Repeated alveolar epithelial cell damage can trigger the occurrence of pulmonary brosis. At the late stage of modeling, ADRP + cells were M2 macrophages, among which a large proportion expressed S100A4 (Fig. 6). This is an important indicator of epithelial mesenchymal transformation. We observed that some adjacent cells of ADRP + macrophages expressed PCNA. These cells were in an active state of proliferation. This suggests that ADRP + macrophages may affect the differentiation and proliferation status of surrounding cells by releasing S100A4. Studies have suggested that S100A4 can be generated and secreted by M2-polarized alveolar macrophages and enhance the proliferation and activation of lung broblasts [21]. S100A4 + cells in IPF lung tissue are distributed in the borderline region between the focal site of myo broblasts and the normal alveolar structure, which is de ned as the pre-brosis activity [21]. These phenomena are mutually con rmed by our observations. It was further suggested that the appearance of ADRP + M2-polarized macrophages was one of the necessary conditions for the occurrence of pulmonary brosis. brosis modeling. c Immunohistochemical staining for ADRP in D7 and D28 pulmonary brosis modeling mice lung. * is P < 0.05; ** is P < 0.01; *** is P < 0.001. Scale bar = 50μm.

Figure 2
Oil red staining positive cells expressed ADRP during pulmonary brosis and distributed in alveolar cavity.
Pulmonary brosis modeling mice lung frozen section (D28) stained with Oil red O, and then eluted (Oil red O elution). ADRP was detected in the same section by immunohistochemical staining (ADRP). CTRL, a parallel immunohistochemical control group, showed that the preceding oil red staining and elution did not affect the immunohistochemical staining. Scale bar = 50μm.

Figure 3
Relationship between ADRP+ cells and different types of macrophages in pulmonary brosis process. Lung specimens were collected at different stages of pulmonary brosis modeling, D7, D14, D21, and D28. a Macrophage-associated molecular markers CD68, D86 and CD206 expression levels were detected by Realtime-PCR in lung tissues of control and pulmonary brosis modeling mice. Immunohistochemical antibody stripping multiple staining was used to detect CD68, CD86 and CD206, colocalization relationship with ADRP in lung. Representative immunohistochemical results were shown in b, c and d respectively and the ratio of ADRP+CD68+, ADRP+CD86+ and ADRP+CD206+ to ADRP+ cells was counted. All experiments repeated in three lung tissue samples. The hollow arrow shows that two molecular markers cannot co-locate in the same cell, while the solid arrow shows that two molecular markers can co-locate in the same cell. NS is P>0.05; * is P<0.05; ** is P<0.01; *** is P<0.001. Scale bar = 50μm.

Figure 4
Distribution of ADRP and α-SMA and N-cadherin in lung of model mice with pulmonary brosis. ADRP, α-SMA and N-cadherin expression pattern were detected by immunohistochemical staining in serial sections of pulmonary brosis tissue prepared on the 21st day after the start of modeling. a Distribution of these target molecules in the brotic lesion area. b Distribution of these target molecules in undamaged areas. Scale bar = 250μm. ADRP+ macrophage aggregation is associated with pulmonary brosis. The experimental mice were intraperitoneally injected with bleomycin for different number of times and fed to 28 days for sampling. a Immunohistochemical staining of α-SMA and ADRP from serial sections of lung tissue revealed that at least three times intraperitoneal injections of bleomycin were required to induce pulmonary brosis. b Changes in the mRNA expression level of CD68, PDGFRα and ADRP gene after different number of intraperitoneal bleomycin injection times. c Immunohistochemical antibody stripping multiple staining of CD206 and ADRP show that three intraperitoneal injections of bleomycin can induce the aggregation of ADRP+ M2 macrophages. * is P<0.05; ** is P<0.01; *** is P<0.001. Scale bar = 50μm. Effect of aggregation of ADRP+M2 macrophages on adjacent peripheral lung tissue cells. The colocalization relationship between ADRP and three target molecules: PCNA, S100A4 and Caspase9 in peripheral lung tissue of D21 brotic mice model was analyzed by immunohistochemical antibody stripping multiple staining technique. a The left immunohistochemical images of PCNA and ADRP showed that ADRP+ cells did not express PCNA. The image on the right is an enlargement of the box on the left to show that most PCNA+ cells are located in the lung interstitial and ADRP+ cells are mainly located in the alveolar cavity. b ADRP+ cells express Caspase9, while many alveolar epithelial cells that do not express ADRP also express Caspase9. c Lots of ADRP+ cells express S100A4 simultaneously. The hollow arrow shows that two molecular markers cannot co-locate in the same cell, while the solid arrow shows that two molecular markers can co-locate in the same cell. Scale bar = 50μm.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.