Status of stem cells in diabetic nephropathy: predictive and preventive potentials

The present study shows reductions in subsets of stem cells in peripheral blood and renal cell preparations of db/db mice, a model of type 2 diabetic nephropathy. We further show increased apoptosis of HSCs, MSCs, and EPCs in cell preparations from kidneys of db/db compared to db/m mice. Collectively, the results suggest that the diabetic milieu exerts detrimental effects on survivability of stem cell subsets in this animal model. Given that stem cells are pivotal players in tissue repair and regeneration following injury, reduction in their numbers (and/or function) impairs the ability of the organism to cope with injury thereby contributing to progressive worsening of renal architecture and dysfunction.

Diabetes is the most common cause of ESRD [2]. Thus, there is an intense interest in slowing/halting the progression of kidney disease and even promoting regression of renal lesions and associated kidney dysfunction. Indeed, remission of the disease and regression of renal lesions can occur in experimental animals and human subjects; regression of glomerular structural changes is associated with remodeling of the glomerular architecture [14]. In specific circumstances, renal injuries can at least partially heal and integrity as well as functionality of the injured portion of the nephron be restored [22–24]. Collectively, these observations indicate that regeneration can occur to some extent in animals and humans. Indeed, the kidney contains a stem/progenitor cell system, defined as the “renopoietic system” dedicated to renal epithelial cell replacement [22]. Further, mobilization of endogenous stem cell reservoir (e.g., bone marrow) and exogenous delivery of stem cell regimens constitute promising venues of addressing diabetic complications such as nephropathy. Importantly, however, this approach must be combined with treatment modalities aimed at achieving strict control of metabolic abnormalities of the disease given the detrimental impact of hyperglycemia on stem cell number and function. This contention is supported by a recent study which investigated the impact of hyperglycemic stress on kidney stem cells which were isolated from the renal papilla showing expression of MSC markers (e.g., N-cadherin, nestin, CD133, CD29, CD90, and CD73) [25]. When these cells were co-cultured with hypoxia-injured renal tubular epithelial cells, they expressed CK18, a marker of mature epithelial cells thereby suggesting that kidney stem cells can differentiate into renal tubular epithelial cells. Importantly, however, culturing kidney stem cells in a high glucose environment impaired their differentiation ability and tolerance to hypoxia. Authors suggest that hyperglycemia compromises the reparative ability of kidney stem cells and could result in decreased ability to recover from injury. The importance of strict metabolic control is also clearly supported by observations that regression of renal lesions in diabetic subjects, following pancreatic transplantation, requires achievement of at least 5 years of normoglycemia [24].

Our observation of reduction in stem cell subsets in the peripheral blood of db/db mice is consistent with other reports indicating that EPCs are reduced in the blood of patients with type 2 diabetes compared to their controls, independent of concomitant risk factors [26]. The reduction in EPCs and associated decreased reparative capacity, in response to endothelial injury, are believed to contribute to a higher risk for cardiovascular disease associated with diabetes mellitus. Consequently, the reduction in EPCs is an important pathogenic factor contributing to microangiopathy which is intimately linked with diabetic complications including nephropathy. Aside from a reduced number, EPCs also show impaired functional features including adhesion, proliferation, and tubulogenesis. Thus, both reduced number and impaired function of EPCs compromise the ability to counter diabetes/hyperglycemia-induced injury [26]. Aside from EPCs, (bone marrow-derived) MSCs play important roles, largely via paracrine mechanism, in repair and regeneration of damaged tissues. Importantly, however, bone marrow-derived MSCs of streptozotocin-induced diabetic rats display impaired proliferation, paracrine release of various factors (e.g., vascular endothelial growth factor), anti-apoptosis, and myogenic differentiation in transplanted tissues [11]. In addition, impairment of bone marrow-derived HSCs is also a feature of diabetes thereby leading to endothelial progenitor cell dysfunction and reduced neovascularization following ischemic insult to the tissue [13].

As alluded earlier, decreased proliferation is believed to contribute to reduced number of stem cells in diabetes mellitus. We now provide evidence that increased apoptosis is also an important contributing factor to decreased number of stem cells in this condition. This is consistent with our recent study indicating marked increase in apoptotic/necrotic cell death in whole kidney cell preparations of db/db than db/m mice, an effect associated with increased GADD153, a marker of increased endoplasmic reticulum (ER) stress response [10]. Aside from ER stress response, hyperglycemia upregulates several pathways (e.g., protein kinase C, polyol pathway, advanced glycation end products, and hexosamine) which along with mitochondrial dysfunction cause increased oxidative stress [25]. In turn, oxidative stress regulates a complex web of signaling pathways which, among other effects, determine cell fate. For example, oxidative stress causes activation of a number of pro-apoptotic kinase signaling intermediates; these include several isoforms of protein kinase C, apoptosis signal-regulating kinase 1, c-Jun-N-terminal kinase and caspase, among others [27]. Thus, hyperglycemia-induced upregulation of pro-apoptotic pathways likely underlies increased cell death not only for the whole kidney as we have shown previously [10] but also in stem cell subsets in the present study in the db/db mice. However, there are very likely to be individual differences in the degree of hyperglycemia-induced upregulation in apoptosis among diabetic patients. This is evidenced by the fact that nephropathy is seen in 25 to 40 % of the diabetics [1]—implying 60 to 75 % have been able to avoid it, at least for some time. Quantitative and qualitative variations abound in biology and medicine, as do all complex adaptive systems. The increase in apoptosis may be an adaptive response to ER stress in some individuals. For these patients, the sum total cost is less to have programmed cell death and avoid spilling DAMP (damage-associated molecular pattern) which can further stimulate the inflammatory responses that are already elevated. What is needed is for us to build and iteratively improve multi-variate high-dimensional computational models based on experimental and clinical data such as provided in this report. It is only with this mechanistic approach that future therapies can be individualized based on serum marker profiles, deep understanding, and logic, rather than experience and hope.