The concept of "stem cell pathology" is to establish the role of the stem cells by exploring their contribution to lesion development. The somatic stem cells are present in the body. Malignant fibrous histiocytoma (MFH; recently named "undifferentiated pleomorphic sarcoma") includes pluripotential undifferentiated mesenchymal stem cells as a cell element. An antibody (A3) generated by using rat MFH cells as the antigen labels somatic stem cells such as bone marrow stem cells and immature endothelial cells and pericytes, as well as immature epithelial cells in epithelialization. By using A3 and other antibodies recognizing somatic stem cells, it is considered that myofibroblasts appearing in rat fibrotic lesions are developed partly from immature hepatic stellate cells in hepatic fibrosis, immature pancreatic stellate cells in pancreatic fibrosis, pericytes/endothelial cells in neovascularization in injured tissues, as well as via the epithelial-mesenchymal transition. These progenitors may be in the stem cell lineage. In this review, the author introduces the histogenesis of MFH and the characteristics of myofibroblasts appearing in fibrosis, based mainly on the author's studies.

Macrophages and myofibroblasts are important in fibrogenesis. The cellular characteristics in pancreatic fibrosis remain to be investigated. Pancreatic fibrosis was induced in F344 rats by a single intravenous injection of dibutyltin dichloride. Histopathologically, the induced pancreatic fibrosis was divided into 3 grades (1+, 2+, and 3+), based on collagen deposition. Immunohistochemically, CD68-expressing M1 macrophages increased with grade and CD163-expressing M2 macrophages also increased later than M1 macrophage appearance. Double immunofluorescence showed that there were macrophages coexpressing CD68 and CD163, suggesting a possible shift from M1 to M2 types; similarly, increased major histocompatibility complex class II- and CD204-expressing macrophages were polarized toward M1 and M2 types, respectively. These findings indicated the participation of M1- and M2-polarized macrophages. Mesenchymal cells staining positive for vimentin, desmin, and α-smooth muscle actin (α-SMA) increased with grade. There were mesenchymal cells coexpressing vimentin/α-SMA, desmin/α-SMA, and glial fibrillary acidic protein (GFAP)/α-SMA; Thy-1-expressing immature mesenchymal cells also increased in fibrotic lesions. Because α-SMA expression is a reliable marker for myofibroblasts, α-SMA-expressing pancreatic myofibroblasts might be originated from GFAP-expressing pancreatic stellate cells or Thy-1-expressing immature mesenchymal cells; the myofibroblasts could simultaneously express cytoskeletal proteins such as vimentin and desmin. The present findings would provide useful information for analyses based on features of macrophages and myofibroblasts in chemically induced pancreatic fibrosis.

Liver cell turnover is very slow, especially compared to intestines and stomach epithelium and hair cells. Since the liver is the main detoxifying organ in the body, it does not come as a surprise that the liver has an unmatched regenerative capacity. After 70% partial hepatectomy, the liver size returns to normal in about two weeks due to replication of differentiated hepatocytes and cholangiocytes. Despite this, liver diseases are regularly encountered in the veterinary clinic. Dogs primarily present with parenchymal pathologies such as hepatitis. The estimated frequency of canine hepatitis depends on the investigated population and accounts for 1%-2% of our university clinic referral population, and up to 12% in a general population. In chronic and severe acute liver disease, the regenerative and replicative capacity of the hepatocytes and/or cholangiocytes falls short and the liver is not restored. In this situation, proliferation of hepatic stem cells or hepatic progenitor cells (HPCs), on histology called the ductular reaction, comes into play to replace the damaged hepatocytes or cholangiocytes. For unknown reasons the ductular reaction is often too little and too late, or differentiation into fully differentiated hepatocytes or cholangiocytes is hampered. In this way, HPCs fail to fully regenerate the liver. The presence and potential of HPCs does, however, provide great prospectives for their use in regenerative strategies. This review highlights the regulation of, and the interaction between, HPCs and other liver cell types and discusses potential regenerative medicine-oriented strategies in canine hepatitis, making use of (liver) stem cells.

When hepatocyte replication during liver disease is insufficient for regeneration, liver progenitor cells (LPCs) are activated. The cells and stroma in the immediate environment of LPCs, together termed the LPC niche, are thought to play an important role in this activation. Among these cells are the hepatic stellate cells (HSCs)/myofibroblasts (MFs).

We assessed the activation of HSC/MFs and LPCs in relation to the histological location and extent of liver disease in immunohistochemically (double) stained serial sections. Markers of HSC/MFs [alpha-smooth muscle actin, glial fibrillary acidic protein (GFAP), neurotrophin 3 and neural-cell adhesion molecule], markers of LPCs (keratin 7 and keratin 19) and a proliferation marker (Ki67) were used. A very relevant spontaneous model to evaluate LPC niche activation in a translational approach seems to be the dog. Therefore, both human and canine liver diseases with different degree of fibrosis and disease activity were included.

In human and canine liver disease, type and extent of LPC niche activation depended on type and severity of disease (P<0.05) and corresponded to the main location of disease. Activated HSCs surrounded the activated LPCs. In chronic hepatitis and non-alcoholic steatohepatitis lobular-type HSCs were activated, while during biliary disease portal/septal MFs were mainly activated. In canine liver, GFAP further presented as an early marker of HSC activation. Activation of the LPCs correlated with disease location and severity (P<0.01), and was inversely related to hepatocyte proliferation, as was previously shown in man.

A shared involvement of HSC/MFs, LPCs and disease severity during hepatic disease processes is shown, which is highly similar in man and dog.

Progressive renal fibrosis is an unwanted and limiting side effect of cancer treatments, whether they are systemic (for example, chemotherapy), local (for example, radiotherapy), or total body irradiation for allogenic bone marrow transplants. The relative roles of macrophages, myofibroblasts, and lymphocytes and the apoptotic deletion of renal functional or inflammatory cell populations in the pathogenesis of renal fibrosis are yet unclear. In this study, rat models of 2 renal cancer treatments: cis-platinum-(II)-diammine dichloride (cisplatin, 6-mg/kg body weight) and radiation (single dose of 20 Gy) were used. Kidneys were analyzed 4 days to 3 months after treatment. The extent of renal fibrosis was compared with number and localization of chronic inflammatory cell populations, cell death (apoptosis and necrosis), and expression and localization of profibrotic growth factors transforming growth factor-beta1 (TGF-beta1) and tumor necrosis factor-alpha (TNF-alpha). The models provided contrasting rates of fibrogenesis: After cisplatin, development of fibrosis was rapid and extensive (up to 50% fibrosis at 3 months); in comparison, radiation-induced fibrosis was slowly progressive (approximately 10% fibrosis at 3 months). The extent of fibrosis was associated spatially and temporally with increasing numbers of myofibroblasts with TGF-beta1 or macrophages with TNF-alpha. Tubular epithelial apoptosis was highest with high TNF-alpha (P<0.05). A significant inverse correlation existed between extent of tubulointerstitial fibrosis and interstitial cell apoptosis for cisplatin and a similar nonsignificant result for radiation (r(2)=0.8671 for cisplatin, P<0.05; r(2)=0.2935 for radiation, NS). The latter result suggests a role for inflammatory cell apoptosis in minimizing development of renal fibrosis.

CD163 is a member of the scavenger receptor cysteine-rich superfamily restricted to the monocyte/macrophage lineage and is thought to be a useful marker for anti-inflammatory or alternatively activated macrophages. In this study we used mass spectrometric analysis to determine that the antigen recognized by the antibody AM-3K, which we previously generated as a tissue macrophage-specific monoclonal antibody, was CD163. An anti-inflammatory subtype of macrophages stimulated by dexamethasone or interleukin-10 showed strong reactivity for AM-3K and increased expression of CD163 mRNA. Immunohistochemical staining of routinely processed pathological specimens revealed that AM-3K recognized a specialized subpopulation of macrophages. In granulomatous diseases such as tuberculosis, sarcoidosis, or foreign body reactions, tissue macrophages around granulomas, but not component cells of the granulomas such as epithelioid cells and multinucleated giant cells, showed positive staining for AM-3K. In atherosclerotic lesions, scattered macrophages in diffuse intimal lesions were strongly positive for AM-3K, whereas foamy macrophages in atheromatous plaques demonstrated only weak staining. We therefore suggest that, in routine pathological specimens, AM-3K is a useful marker for anti-inflammatory macrophages because these cells can be distinguished from inflammatory or classically activated macrophages. Because AM-3K cross-reacts with macrophage subpopulations in different animal species including rats, guinea pigs, rabbits, cats, dogs, goats, pigs, bovine species, horses, monkeys, and cetaceans, it will have wide application for detection of CD163 in various animals.

Myofibroblasts play an important role in chronic renal interstitial fibrosis. However, the origin and developmental mechanisms remain to be elucidated. The myofibroblasts may express various cytoskeletons during the development. Immunoexpressions of vimentin, desmin and alpha-smooth muscle actin (alpha-SMA) were analyzed using experimentally (cisplatin and unilateral ureteral obstruction) induced rat and spontaneous canine fibrotic kidneys or kidney-related cell lines incubated with transforming growth factor-beta1 (TGF-beta1), platelet-derived growth factor-BB (PDGF-BB) or their combination at various concentrations. In rat fibrotic kidneys, both renal epithelia and interstitial cells showed positive reactions to alpha-SMA and vimentin, supporting epithelial-mesenchymal transition (EMT) theory; however, renal epithelia did not react to desmin, though interstitial cells were reactive. Renal epithelia in canine fibrotic kidneys did not show a positive reaction to alpha-SMA, whereas interstitial cells reacted strongly to alpha-SMA; conversely, renal epithelia reacted strongly to desmin, but interstitial cells did not; vimentin expression was infrequently seen in renal epithelia and interstitial cells of canine kidneys. Exposure of TGF-beta1 to porcine renal epithelial cells (LLC-PK1), rat renal interstitial cells (NRK-49F), and rat immature mesenchymal cells (MT-9) dose-dependently increased selectively alpha-SMA-positive cell numbers. Moreover, PDGF-BB exhibited an additive effect on TGF-beta1-induced alpha-SMA expression in these cell lines when simultaneously added. alpha-SMA was the most plastic cytoskeleton under fibrogenic stimuli. This study shows that there are interspecies differences in cytoskeletal immunoexpressions of renal epithelia or interstitial cells between rat and canine fibrotic kidneys, and that the derivation of renal myofibroblasts may be heterogeneous, such as renal epithelia, interstitial cells or immature mesenchymal cells.

Macrophages may play a role in fibrogenesis. The kinetics and distribution of different macrophage populations were investigated immunohistochemically in hepatic lesions following acute hepatocyte injury induced in F344 rats by a single injection of thioacetamide (TAA) (300 mg/kg body weight, intraperitoneally). Hepatocyte degeneration or necrosis induced by TAA occurred mainly in the perivenular areas of hepatic lobules as early as post-injection (PI) days 1 and 3; fibrotic lesion development began in the damaged areas on day 1, and peaked on day 5; thereafter (PI days 7 and 10), the fibrotic areas decreased and were replaced by regenerated hepatocytes on PI days 15 and 20, indicating a remodelling process. In this rat model, the number of macrophages reacting with ED1 antibody (specific for exudate macrophages), ED2 (recognizing cell membrane antigens of resident macrophages, including Kupffer cells) and OX6 (recognizing MHC class II antigens expressed in antigen-presenting macrophages and dendritic cells) began to increase on PI day 1, peaking on PI day 3. The numbers gradually decreased on PI days 5 and 7; however, the statistically significant increase was maintained in respect of ED1-positive cells up to PI day 20, whereas no significant increase in ED2- and OX6-positive cells remained from PI day 10 onwards. Interestingly, of the ED1-, ED2- and OX6-positive cells, the OX6-positive cells were the least numerous. ED1- and OX6-positive cells appeared exclusively in the injured perivenular areas, whereas ED2-positive cells were present mainly in the mid-zonal areas and in smaller numbers in the perivenular areas. These findings indicated differences in kinetics and distribution between macrophage populations appearing in hepatic fibrosis. In addition, RT-PCR revealed that mRNA expression of osteopontin, a factor for induction and maintenance of macrophages in inflammation, was markedly increased on PI days 5, 7 and 10, suggesting a role in the pathogenesis of hepatic fibrosis.

The signature lesion of autoantibody-associated congenital heart block (CHB) is fibrosis of the conducting tissue. To date, participation of myofibroblasts in the cascade to injury has been unexplored. The importance of myofibroblast/macrophage cross-talk is demonstrated by the novel finding of these cell types in the heart of a neonate dying of CHB. This clue to pathogenesis prompted consideration of the mechanism by which maternal anti-SSA/Ro-SSB/La Abs initiate an inflammatory response and promote fibrosis. Isolated cardiocytes from 16-24 wk abortuses were rendered apoptotic by exposure to poly (2-) hydroxyethylmethacrylate; flow cytometry confirmed surface expression of Ro/La. Apoptotic cardiocytes were incubated with affinity-purified Abs to 52 and 60 kDa Ro from CHB mothers (opsonized) or IgG fractions from healthy donors (nonopsonized). Macrophages cultured with opsonized apoptotic cardiocytes expressed proinflammatory markers, supported by a three-fold increase in active alpha(V)beta(3) integrin. Fetal cardiac fibroblasts exposed to supernatants obtained from macrophages incubated with opsonized apoptotic cardiocytes (but not nonopsonized) dramatically increased expression of the myofibroblast marker alpha-smooth muscle actin (SMAc). The "opsonized" supernatant reversed an inhibitory effect of the "nonopsonized" supernatant on proliferation of fibroblasts (120 vs 69%, p < 0.05). Parallel experiments examined the effects of two cytokines and their neutralizing Abs on fibroblasts. TGFbeta1 increased SMAc staining but decreased proliferation. TNF-alpha did not affect either readout. Addition of anti-TGFbeta1 Abs to the "opsonized" supernatant blocked SMAc expression but increased proliferation, while anti-TNF-alpha blocking Abs had no effects. These data suggest that transdifferentiation of cardiac fibroblasts to a scarring phenotype is a pathologic process initiated by maternal Abs.