Histopathology Unit, Cancer Research UK London Research Institute, London, United Kingdom
Histopathology Unit, Cancer Research UK London Research Institute, 44 Lincoln's Inn Fields, London, WC2A 3PX, United Kingdom and Department of Gastroenterology, Faculty of Medicine, Imperial College London, Hammersmith Campus, Du Cane Road, London W12 0NN, United Kingdom. Telephone: 44-020-7269-3245; Fax: 44-020-7269-3491
Myofibroblasts are ubiquitous cells with features of both fibroblasts and smooth muscle cells. We suggest that the bone marrow can contribute to myofibroblast populations in a variety of tissues and that this is exacerbated by injury. To assess this, female mice were transplanted with male bone marrow and the male cells were tracked throughout the body and identified as myofibroblasts. Skin wounding and paracetamol administration were used to assess whether myofibroblast engraftment was modulated by damage. Following radiation injury, a proportion of myofibroblasts in the lung, stomach, esophagus, skin, kidney, and adrenal capsule were bone-marrow derived. In the lung, there was significantly greater engraftment following paracetamol administration (17% versus 41% p < 0.005). Bone-marrow-derived fibroblasts were also found. We suggest that bone marrow contributes to a circulating population of cells and, in the context of injury, these cells are recruited and contribute to tissue repair.
Cells from the bone marrow have now been shown to have the capacity to engraft into several organs and differentiate into many cell types (stem cell plasticity) [1–10], and even apparently fully differentiated cells can switch fate and transdifferentiate into other distinct cell types . Adult bone marrow shows remarkable plasticity, and work by Krause et al. has shown multilineage epithelial engraftment by a single highly purified transplanted bone-marrow-derived stem cell . Further work by Körbling et al. in humans has shown that donated peripheral blood mobilized hematopoietic adult stem cells can also transdifferentiate, in that instance, into mature hepatocytes as well as epithelial cells of the skin and gut .
Myofibroblasts are widely distributed cells with features of both fibroblasts and smooth muscle cells. They are important coordinating cells that have a significant influence on their environment by virtue of the receptors they possess and the cell signals they produce. Their functions include roles in growth, differentiation, development, and healing and the inflammatory response. They have the capacity to produce extracellular matrix and, indeed, overactivation of myofibroblasts may be the underlying process in fibrosis, scarring, and many fibrotic diseases (e.g., pulmonary fibrosis [14, 15] and liver fibrosis [16, 17]).
We recently reported that pericryptal myofibroblasts in the mouse and human small intestine are bone marrow derived . By transplanting male whole bone marrow into female mice, we were able to track the male-donor-derived cells in the female recipient by detecting the Y chromosome using in situ hybridization. Myofibroblasts in the gut were identified on the basis of their morphologies and expression of alpha-smooth muscle actin (α-SMA). Employing the same experimental technique , we have now established that bone-marrow-derived myofibroblasts are present in multiple organs and, in certain circumstances, fibroblasts can also be bone marrow derived.
Wang et al. illustrated the crucial role of damage, such as lethal irradiation, to act as a selection pressure and encourage engraftment of bone-marrow-derived cells after sex-mismatched bone marrow transplantation . Here, we use other damage modalities—skin wounding and paracetamol administration—to exert a similar selection pressure and to assess recruitment of bone-marrow-derived myofibroblasts. Krause et al. described epithelial engraftment after hematopoietic stem cell transplantation , but here we have focused on myofibroblasts as cells with a pivotal role in the stem cell niche and, of course, fibrosis. On the basis of this work, we propose the hypothesis that circulating bone-marrow-derived cells give rise to myofibroblasts and fibroblasts, that this engraftment occurs in multiple organs, and that it is exacerbated by injury. This provides yet more evidence for the plasticity of adult bone marrow and opens up new avenues of therapeutic potential.
Materials and Methods
All animal procedures were carried out under British Home Office procedural and ethical guidelines. Young adult female recipient mice (C57/black) underwent whole body γ-irradiation, with 12 Gy in a divided dose 3 hours apart, to ablate their bone marrow. This was followed immediately by tail vein injection of one million male wild-type whole bone marrow cells (three male C57/black donor mice supplied bone marrow for 10 recipient female mice). The mice were housed in sterile conditions. Additional injury was delivered to a proportion of the mice to assess whether this influenced the amount or distribution of engraftment. Five mice were treated with paracetamol (400 mg/kg i.p.) at 5 and 8 weeks after bone marrow transplantation. The animals were sacrificed at intervals up to 13 weeks following bone marrow transplant, and their organs were fixed in neutral-buffered formalin before being embedded in paraffin wax. Skin injury was induced in a separate group of five mice (no paracetamol) with a punch biopsy 10 weeks after sex-mismatched bone marrow transplantation. Skin was harvested at 4 and 7 days postwounding and embedded in paraffin wax as above.
Immunohistochemical Analyses and Y Chromosome Detection
To identify the origin of myofibroblasts in multiple organs, immunohistochemistry was combined with in situ hybridization for the Y chromosome.
To identify myofibroblasts in mouse tissue sections, we immunostained for α-SMA (mouse monoclonal Clone 1A4, A-2547; Sigma; Poole, UK; http://www.sigmaaldrich.com) or vimentin (Clone 3B4, DAKO M7020; DAKO; Ely, UK; http://us.dakocytomation.com) prior to in situ hybridization for the Y chromosome. Four-micron thick sections were dewaxed and incubated with hydrogen peroxide (2.4 ml 30%) in methanol (400 ml) to block endogenous peroxidases and taken through graded alcohols to phosphate-buffered saline (PBS). All tissues were incubated for 3 minutes in acetic acid (20%) in methanol to block endogenous alkaline phosphatase. Slides were preincubated in normal rabbit serum (DAKO D0396) at 1/25 dilution in PBS for 10 minutes. The slides were then incubated in primary antibody (SMA) at a dilution of 1/4,000 in PBS for 35 minutes. The secondary antibody was a biotinylated rabbit anti-mouse (DAKO E0354) and was applied for 35 minutes. A tertiary layer of streptavidin-alkaline phosphatase (DAKO D0396) diluted to 1/50 in PBS for 35 minutes followed this. Sections were washed in PBS between each antibody layer, and Vector Red substrate (SK 5100; Vector Laboratories; Peterborough, UK; http://www.vectorlabs.com) was applied for 15 minutes at room temperature. Sections were again washed in PBS prior to the in situ hybridization protocol.
In Situ Hybridization
Sections were incubated in 1 M sodium thiocyanate for 10 minutes at 80°C, washed in PBS, then digested in pepsin (0.4% w/v) in 0.1 M HCl at 37°C for varying times dependent upon the tissue being studied. The protease was quenched in glycine (0.2% w/v) in double-concentration PBS, and the sections were then rinsed in PBS, postfixed in paraformaldehyde (4% w/v) in PBS, dehydrated through graded alcohols, and air dried. A fluorescein-isothiocyanate-labeled Y chromosome paint (Star-FISH; Cambio; Cambridge, UK) was used in the supplier's hybridization mix. The probe mixture was added to the sections, sealed under glass with rubber cement, heated to 60°C for 10 minutes, and incubated overnight at 37°C. The slides were then washed in formamide (50% w/v)/2× standard saline citrate (SSC) at 37°C, then washed with 2× SSC and then 4× SSC/Tween-20 (0.05% w/v) at 37°C. All slides were then washed with PBS and incubated with 1/25 peroxidase-conjugated antifluorescein antibody (150 U/ml; Boehringer Mannheim; Indianapolis, IN; http://www.roche-applied-science.com) for 60 minutes at room temperature. Slides were developed in 3,3′-diaminobenzidine (0.005 g in 10 ml PBS) plus hydrogen peroxide (20 μl), counterstained with hematoxylin, and mounted.
To assess the degree of engraftment of bone-marrow-derived cells in the lung with and without paracetamol, 1,000 myofibroblasts were counted in each group. The percentage of donor-derived myofibroblasts within each group was noted and the mean was calculated. The mean percentage of bone-marrow-derived myofibroblasts of the paracetamol-treated group was then compared with the non-paracetamol-treated group using a two-tailed t-test. The percentages of donor-derived myofibroblasts were also calculated in the stomach and kidney. To assess the degree of engraftment in the skin, the number of donor-derived cells per high-power field (40× objective) was noted, and the level of engraftment with and without wounding was compared using a two-tailed t-test. Sectioning tissue results in division of the nucleus and, therefore, detection of the Y chromosome is not expected in every Y-chromosome-containing cell. To correct for this, the number of Y-chromosome-positive myofibroblasts in male murine tissue was determined, and the cell counts in the transplanted mice were corrected by this factor.
Myofibroblasts were identified on the basis of their morphologies and positive immunoreactivity for α-SMA. Y-chromosome-containing myofibroblasts were seen in the small intestine and colon, as previously reported . In addition, they were also identified in the lung, adrenal capsule, kidney, stomach, and skin (Figs. 1 and 2, Figure 2.). In the lung, donor-derived myofibroblasts were found close to the major airways and vessels as well as distributed in the lung parenchyma (Fig. 1A). Higher levels of engraftment were seen in the lung specimens of the paracetamol-treated animals: 41% (standard deviation [SD] = 14%) of myofibroblasts were found to be donor derived in the lung following administration of paracetamol 5 and 8 weeks after male bone marrow transplantation compared with 17% (SD = 10%) of myofibroblasts without this treatment (p < 0.005). These values were corrected by a factor of 0.57 (standard error = 0.02), which was the proportion of Y-chromosome-positive myofibroblasts in the male mouse lung.
In the adrenal gland, α-SMA-positive cells were seen in the adrenal capsule (Fig. 1B). These cells were clearly demonstrated, although they were a rare occurrence, with only one to two donor-derived cells being seen in each adrenal cross-section. In the kidney, Y-chromosome-positive, α-SMA-positive cells were seen in the interstitium (Fig. 1C). In the stomach, 64% (SD = 9%) of myofibroblasts were seen to be donor derived (Fig. 1D). The degree of engraftment is documented in Table 1.
Table Table 1.. Degree of myofibroblast engraftment in various tissues
In the lung and kidney, the degree of engraftment with and without paracetamol damage and, in the skin, the degree of engraftment with and without wounding were compared using a two-tailed t-test. In the stomach, the degree of engraftment is shown after bone marrow transplantation without additional injury.
*It was impossible to accurately determine the total number of myofibroblasts in wounded skin; thus, the number of Y-chromosome- and α-SMA positive (donor) cells were counted per high-power field (40× objective) for a total of 67 fields.
†NA = not applicable. No myofibroblasts were found in normal skin.
Time after bone marrow transplant (weeks)
Mean % ± SD donor-derived myofibroblasts with wounding
Mean % ± SD donor-derived myofibroblasts without wounding
41 (± 14)
17 (± 17)
24 (± 7)
20 (± 6)
64 (± 9)
4 (± 3.4)
0.07 (± 0.3)
In one animal, a small abscess cavity was found adjacent to the intestine. Interestingly, the wall of the abscess contained multiple donor-derived Y-chromosome-positive cells (Fig. 1E and 1F). These male cells had an elongated fibroblastic morphology and were negative for α-SMA (Fig. 1E) but positive for vimentin (Fig. 1F) in keeping with their fibroblastic characteristics.
In the skin after wounding there was marked α-SMA positivity at the point of injury (Fig. 2A) and, within this region, donor-derived α-SMA-positive cells were found (Fig. 2A, inset). Further, donor-derived myofibroblasts were found nearby (Fig. 2B). Even without wounding, a few donor-derived myofibroblasts were found in close proximity to the hair follicle (Fig. 2C). The degree of lymphoid chimerism was not assessed in any of the transplanted mice.
There is now increasing evidence to suggest that there is a far greater degree of plasticity in adult stem cells than previously thought . The bone marrow is of particular interest, and evidence has already been reported to support its plasticity in the liver [1, 2, 4], muscle , heart , brain , and kidney . In these tissues, engraftment of bone-marrow-derived cells has been documented by tracking the Y chromosome in sex-mismatched bone marrow transplant recipients. Bone marrow is easily accessible and contains both mesenchymal and hematopoietic stem cells. Bone marrow transplantation is a technique already used extensively in humans, and individual human case studies have already supported laboratory findings [13, 18, 22]. Indeed, recent work has shown that, following sex-mismatched conventional whole bone marrow transplantation, donor-derived osteoblasts and osteocytes were found in trephine biopsy samples .
Myofibroblasts are ubiquitous cells whose absence in development is associated with grossly disordered structure, for example, in the gut  and lung . Myofibroblasts are classified on the basis of their cytoskeletal elements—the presence and/or absence of microfilaments (α-SMA) and intermediate filaments (desmin and vimentin). Myofibroblasts produce a multitude of chemokines, cytokines, and growth factors. This allows them to influence their environment: for example, they have a coordinating role in the inflammatory response (e.g., via interleukin-6), while the growth factors they produce (e.g., hepatocyte growth factor, keratinocyte growth factor, transforming growth factor [TGF]-α, and TGF-β) can promote the differentiation and proliferation of surrounding parenchymal epithelial cells.
Myofibroblasts are contractile cells and begin the process of wound healing by reducing the area of denuded basement membrane in injury. This ability is only one aspect of their role in the healing process; myofibroblasts also produce extracellular matrix molecules such as collagens I-VI, proteoglycans, and matrix-modifying proteins. This emphasizes the importance of their role in healing and repair. However, the healing process can become overactivated, and excess production of these matrix molecules can result in scarring and, in extreme cases, fibrosis. Overactivation of myofibroblasts has already been shown to be associated with disease in many organs in humans [26, 27], while myofibroblast dysfunction or absence has also been shown to result in disease in mice .
We reported previously that approximately 50% of pericryptal myofibroblasts in the mouse and human small intestine are bone marrow derived . We report here that myofibroblasts in multiple organs can be similarly bone marrow derived. We have demonstrated bone-marrow-derived myofibroblasts in the stomach, lung, skin, kidney, adrenal gland, colon, and small intestine. This suggests that the bone marrow can potentially contribute to the turnover of myofibroblasts throughout the body. However, this occurred in the context of lethal irradiation and bone marrow transplantation and, therefore, a degree of radiation injury to all organs.
Abe et al.  showed that peripheral blood contains a population of circulating fibrocytes and that these fibrocytes migrate to areas of injury in the skin of both mice and humans. We have suggested that this process may also be operative in the gut  and here we propose that these cells can be derived from the bone marrow and contribute to this systemic phenomenon. Krause et al.  showed that multiple organ engraftment of epithelial cells could occur when a single purified bone-marrow-derived hematopoietic stem cell was transplanted into irradiated mice, but did not report observing donor-derived myofibroblasts. It is our hypothesis that circulating fibrocytes are derived from transplanted bone marrow mesenchymal stem cells/stromal cells and that these circulating fibrocytes finally engraft to injured tissue throughout the body.
Why is this important? It is already clear that myofibroblasts are important cells in organ development and repair and that dysfunction of myofibroblasts can result in disease. Bone-marrow-derived myofibroblasts are deeply engrafted into their host tissue; thus, bone marrow transplantation may be used as a vehicle for gene therapy and, in disease, areas of increased myofibroblast activity may be targeted. Moreover, where there is an absence or hypofunction of myofibroblasts, for example, in the platelet-derived growth factor-α  knockout mouse, which has gastrointestinal, renal, and cardiovascular abnormalities , bone marrow transplantation might rectify this deficiency.
There are several theories of cytogenesis in the adrenal cortex, one of which is the migration theory, whereby cells are born in the zona glomerulosa and migrate inward . Originally this theory stated that the cells that migrated took origin in the capsule, and we have seen Y-chromosome-/α-SMA-positive cells in the capsule and cells that are Y-chromosome positive but α-SMA negative. Thus, it is conceivable that the bone marrow contributes to cells in the adrenal capsule and, in turn, the capsule makes a contribution to the adrenal cortex.
Acute lung injury and pneumonitis have been reported in patients either treated with or overdosed on paracetamol [31, 32]. Paracetamol is normally conjugated in the liver to give glucuronide or sulphate metabolites. When these pathways become saturated, paracetamol is then metabolized via the cytochrome P-450 (CYP) system to give the toxic metabolite N-acetyl-p-benzoquinone. This toxic compound is then inactivated by conjugation with glutathione. However, once glutathione stores are depleted, the toxic metabolite accumulates, ultimately resulting in necrosis. In overdosed mice, paracetamol binds to hepatocytes with subsequent necrosis; this covalent binding colocalizes with that of CYP 2E1. This also occurs in the renal proximal tubules, olfactory epithelium, and bronchiolar epithelium . Further studies in rats show that, as expected in paracetamol toxicity, glutathione stores in the lung are depleted . Cultured human fibroblasts have a higher proportion of DNA single-strand breaks when irradiated in the presence of paracetamol than when it is absent . In our study, a group of animals were treated with paracetamol following the irradiation required for successful bone marrow transplantation. Consistent with the studies suggesting extra lung damage and reduced antioxidant capacity, a doubling of the levels of bone-marrow-derived myofibroblasts was seen in the lungs of those animals treated with paracetamol.
We show here that bone-marrow-derived cells have the capacity to engraft into multiple organs and contribute to the myofibroblast population in these organs. We hypothesize that bone-marrow-derived stem cells are the source of a circulating population of fibrocytes that is recruited to areas of damage. We suggest that abnormalities in this axis may begin to explain fibrotic disease, particularly multisystem disorders, and postulate that an understanding of this axis may facilitate the development of new therapies.
This research was supported by Cancer Research UK.