Stem Cell–Based Therapy for Glomerular Diseases: An Evolving Concept

Alport syndrome is a genetic disease that affects kidneys, ears, and eyes as a result of mutations in basement membrane collagen genes COL4A3, COL4A4, and COL4A5, encoding the α3, α4, and α5 chains of type IV collagen chains.¹ Deletion of COL4A3 gene in mice (α3KO mice) leads to a phenotype that resembles human Alport renal disease. In these mice, deletion of α3 chain results in defective assembly of α3/α4/α5 type IV collagen triple helical protomers in the glomerular basement membrane (GBM). The GBM type IV collagen composition in the α3KO mouse is composed mostly of α1/α2/α1 protomers, an embryonic GBM composition, and, as mice develop into adulthood, the GBM presumably cannot withstand increasing glomerular filtration pressure and undergoes rapid endoproteolysis/turnover to generate the characteristic GBM breaks as observed by electron microscopy. As a result, α3KO mice develop progressive glomerulonephritis similar to those observed in patients with Alport syndrome and eventually die of renal failure.

Interestingly, the strain background of mice confers significant differences in the rate of disease progression, with much longer survival observed in α3KO mice on a C57Bl/6 genetic background, which live up to 25 to 30 wk, in comparison with the 129sv genetic background, which live only 12 to 13 wk. Recent studies suggested a strain-dependent GBM deposition of α5/α6/α5 type IV collagen triple helical protomers in the α3KO on a C57Bl/6 background yet with almost undetectable levels in the α3KO on a 129sv background.² Such difference in GBM composition may offer better survival advantage, likely as a result of slower rate of GBM damage because of the presence of type IV collagen molecules (α5 and α6) with better cross-linking capacity.³ These findings not only highlight the role of type IV collagen specificity in maintaining GBM structure and function but also substantiate the notion that a potential improvement in GBM architecture and renal function can be achieved in Alport renal disease by altering the type IV collagen GBM composition to more stable molecules such as the α3 through α6 chains.^2,3 Indeed, Sugimoto et al.⁴ along with Prodromidi et al.⁵ provided evidence for the therapeutic benefit of bone marrow transplantation (BMT) in irradiated α3KO mice on the C57Bl/6 background. In contrast to irradiated α3KO mice that received α3KO bone marrow, the irradiated α3KO mice that received wild-type bone marrow revealed significant improvement in GBM architecture associated with de novo expression of α3 chain mRNA and protein. Furthermore, GBM protein analyses demonstrated the newly deposited α3 chains likely assemble into viable triple helical protomers. These mice also revealed statically significant improvement in renal function. In the study by Prodromidi et al.,⁵ irradiated α3KO mice that received α3KO bone marrow did not show improved renal functions after BMT and died from renal failure, similar to the study by Sugimoto et al.⁴

Katayama et al.⁶ in this issue of JASN, however, challenge these findings and argue that irradiation alone prolongs the survival of α3KO mice (129sv background) without any need for BMT. This interesting observation is a welcome addition to the debate of whether stem cells are of any therapeutic use in the setting of Alport renal disease. Such studies are essential to fuel collegial discussion about the strategies and approaches to the study of stem cell biology as it relates to the kidney and testing the utility of cell-based therapy for chronic kidney diseases.

Katayama et al. demonstrate that irradiation alone increases the survival of α3KO mice to 22 wk, whereas the control mice normally die from renal failure at approximately 13 wk of age.⁶ Although urine albumin measurements were not provided, the authors suggest that irradiation dosages of 3 and 6 Gy alone to 3-wk-old mice significantly improve serum creatinine and blood urea nitrogen renal function parameters, yet interstitial fibrosis remains unchanged when compared with control mice. Interestingly, improvement in interstitial fibrosis is observed only in the cohort that received BMT after irradiation, but, curiously, such substantial difference in fibrosis did not lead to a survival benefit when compared with α3KO with irradiation alone. The authors also indicate that they fail to detect de novo expression of the α3 chain of IV collagen protein by immunohistochemistry and Western blot in irradiated α3KO mice that were rescued with wild-type bone marrow but did detect weak expression by reverse transcriptase–PCR in irradiated α3KO mice that were rescued with wild-type bone marrow. On the basis of these results, Katayama et al. suggest that an increase in overall survival in the irradiated α3KO mice with BMT, regardless of the donor's genotype, could be due to an increase in α6 (IV) collagen expression in the GBM. Such a suggestion argues that radiation alone can modulate the expression of type IV collagen genes, but, unfortunately, no such evidence was provided for α3KO mice receiving irradiation alone.

A deeper analysis of this interesting and provocative report by Katayama et al. offers some possible insights into the different outcome of cell-based therapy for α3KO Alport mice. First, the mice used in the study by Katayama et al.⁶ are on the 129sv background, which has a significantly different rate of disease progression and kinetics of GBM damage when compared with the mice used in the studies by Sugimoto et al.⁴ and Prodromidi et al.⁵ (C57Bl/6 background). The irradiation (3/6 Gy) and BMT was performed at 3 wk of age,⁶ when compared with 10 Gy of irradiation and BMT at 8 wk of age in the study by Sugimoto et al.⁴ Notwithstanding the strain differences between these two studies, it is likely that at 3 wk, α3KO mice on the 129sv background are normal without significant GBM defects and proteinuria, as reported by Cosgrove and colleagues.^7,8 Conversely, α3KO mice on the C57Bl/6 background at 8 wk of age have significant defects in GBM with proteinuria, a likely prerequisite for successful cell-based therapy. Furthermore, Borza and colleagues² reported the presence of α5 and α6 chains of type IV collagen in the GBM of α3KO on the C57Bl/6 background but not in mice on the 129sv background. A suggestion is made that C57Bl/6 mice live much longer as a result of incorporation of α5- and α6-containing protomers into the glomerulus, when compared with 129sv mice. Katayama et al. do not cite this reference accurately and suggest that the GBM of α3KO mice on the 129sv background contain α5 and α6, contrary to the finding of Borza and colleagues,² which brings up the issue of how Katayama et al. could detect α5 and α6 in normal α3KO mice on the 129sv background when both studies use mice and antibodies from the same source.

Careful examination of immunohistochemistry and Western blot images in the study by Katayama et al. reveals a likely faint labeling for α3 chain, if stringently evaluated, as also reported in the studies by Sugimoto et al. and Prodromidi et al.. In those studies, it was made clear that the expression of α3 chain was quite mild compared with the α5 chain; therefore, an increase in the expression of α5 and α6 chains along with a very mild increase in α3 chain could offer an explanation for improved glomerular histology, a decrease in interstitial fibrosis, and improved renal function in the study by Katayama et al.. This assessment could have been possible if the authors had provided histology and Western blots of rescued mice at 18 to 22 wk of age along with an assessment of mice at 7 to 11 wk of age.

Katayama et al. demonstrate that irradiation alone at 3 wk of age offers survival benefit to α3KO mice on the 129sv background. This effect could be unrelated to the benefit observed as a result of cell-based therapy performed after disease induction and GBM defects accumulation (and more clinically relevant), as seen in the studies by Sugimoto et al.⁴ and Prodromidi et al.⁵ In this regard, renal disease associated with α3KO mice has a dominant inflammatory component, and monocytes and cell-mediated immunity play a key role in disease induction and progression independent of structural defects in the GBM^7,9,10,11; therefore, as also reported for other mouse models of inflammatory renal disease,¹² irradiation has a significant impact in attenuating rapid disease progression, as also observed in various human renal disease with the use of immunosuppressants (irrespective of initiating cause). The study by Katayama et al. once again validates such nonselective therapy options for inflammatory kidney disease. Nevertheless, this study opens up a healthy debate among renal researchers regarding how best to perform preclinical studies to assess the therapeutic potential of stem cells for chronic renal diseases. Of course, it is clear that better models and even better molecular readouts are required before we prematurely shut our minds to this possibility. Importantly, stem cell experiments with mouse models of induced acute kidney disease such as ischemia-reperfusion injury cannot become authoritative examples of feasibility of stem cell therapy for chronic renal diseases that are inherently irreversible by nature. The renal community looks forward to an exciting discussion as novel cell-based therapy strategies are explored to help our patients.

• © 2008 American Society of Nephrology