Souma T, Yamazaki S, Moriguchi T, et al.
Plasticity of renal erythropoietin-producing cells governs fibrosis.
J Am Soc Nephrol.
2013;24:1599-1616.

Chronic kidney disease (CKD) is characterized by erythropoietin (EPO)- deficiency anemia, renal fibrosis, and scarring. A large majority of circulating EPO is produced by a population of interstitial fibroblastlike cells in the renal cortex and outer medulla. In healthy individuals, the number of these renal EPO-producing cells (REPs) is low, but the number increases when decreased oxygen delivery due to hypoxia triggers hypoxiainducible factor-mediated EPO production. In CKD, anemia develops as the number of REPs and the resultant concentration of circulating EPO are decreased despite the tissue hypoxia caused by anemia. Thus, EPO deficiency is the major cause of anemia in renal failure.1  This EPO deficit in CKD may be due to loss of REPs, locally increased oxygen levels in the microenvironment of the REPs, or an altered relationship between local oxygenation and EPO production by REPs.1 

In CKD, renal fibrosis and scarring are mediated by myofibroblasts, a subset of mesenchymal cells that produce collagen and can be identified by their expression of desmin and α-smooth muscle actin (α-SMA). In mice with renal inflammation that is induced by unilateral ureteral obstruction (UUO), the majority of kidney myofibroblasts proliferate in response to inflammatory cytokines and are derived from cells that reside in the kidney, but are neither epithelial cells nor vascular pericytes.2  A minority of myofibroblasts in this model are derived from differentiation of marrow-based progenitors that do not proliferate in renal tissue. The origin of the intrarenal cells that divide and differentiate into myofibroblasts is unknown, but REPs are likely candidates, because, with inflammatory stimuli, they show decreased evidence of EPO expression while desmin3  and α-SMA4  expressions are induced.

Using mice in which REPs can be both identified and their state of EPO production determined, Tomokazu Souma et al. demonstrate that renal inflammation from UUO transforms REPs into myofibroblasts that do not produce EPO, but rather mediate kidney fibrosis. Similar results were found in these mice using unilateral ischemia reperfusion injury and protein overload nephropathy models. This transformation was associated with both cell proliferation and morphologic transformation. Angular EPO-producing REPs that bridged capillaries and proximal tubules decreased, while an increase was observed in stellate REPs that no longer produced EPO and had less extension and more rounded nuclei characteristic of myofibroblasts. Although UUO induced the majority of REPs to assume the characteristics of myofibroblasts, expressing α-SMA and collagen, when the UUO was relieved after two days, the morphologic and biochemical changes were completely reversed over the ensuing 12 days. The recovery of EPO production and loss of collagen expression was hastened when the mice that had been subjected to short-term UUO were treated with dexamethasone. The transformation of EPO-producing REPs to myofibroblasts by UUO appeared to be induced by inflammatory cytokines produced in the REPs themselves, as expression of TGFβ1 and NFκB increased, and their respective signaling factors, Smad 2/3 and phosphorylated p65, showed nuclear accumulation. The recapitulation of reversible transformation of REPs to myofibroblasts, including the suppression of EPO expression following lipopolysaccharide (LPS) injection, and the attenuation of this effect by dexamethasone, suggested that the NFκB pathway had an important role in suppression of EPO production during transformation of the REPs. The inability of TGFβ1 injections to suppress EPO expression in REPs suggested that the major effect of this inflammatory cytokine was on driving fibrogenic activity in the myofibroblasts.

This study by Souma and colleagues demonstrates that, in the short term, the fibroblast-like interstitial cells that produce the majority of EPO in the body are not destroyed as a consequence of renal inflammation, but rather they are transformed into myofibroblasts that cause fibrosis and scarring in CKD. The duration of reversibility and corticosteroid responsiveness of the transformed REPs is unknown. Indeed, these features may vary depending upon the cause of renal disease, but the characteristics of reversibility and steroid responsiveness suggest an opportunity for therapeutic intervention in the early stages of renal injury. Cytokine suppression of renal EPO production is not only confirmed by this study, but the experimental evidence supports the hypothesis that REPs are a source of the cytokines. Interruption of this cytokine-mediated transformation process may not only ameliorate EPO deficiency in patients with CKD, but also inhibit the progression to end-stage renal disease.

1.
Nangaku M, Eckardt KU.
Pathogenesis of renal anemia.
Semin Nephrol.
2006;26:261-268.
http://www.ncbi.nlm.nih.gov/pubmed/16949463
2.
LeBleu VS, Taduri G, O’Connell J, et al.
Origin and function of myofibroblasts in kidney fibrosis.
Nat Med.
2013;19:1047-53
http://www.ncbi.nlm.nih.gov/pubmed/23817022
3.
Maxwell PH, Ferguson DJ, Nicholls LG, et al.
The interstitial response to renal injury: fibroblast-like cells show phenotypic changes and have reduced potential for erythropoietin gene expression.
Kidney Int.
1997;52:715-724.
http://www.ncbi.nlm.nih.gov/pubmed/9291192
4.
Asada N, Takase M, Nakamura J, et al.
Dysfunction of fibroblasts of extrarenal origin underlies renal fibrosis and renal anemia in mice.
J Clin Invest.
2011;121:3981-3990.
http://www.ncbi.nlm.nih.gov/pubmed/21911936

Competing Interests

Dr. Koury indicated no relevant conflicts of interest.