Ang II appears to play a critical role in the development of glomerular capillary hypertension following kidney mass ablation and may also contribute to changes in K _f . The acute infusion of Ang II in normal rats results in a rise in P _GC due to a greater increase in efferent than afferent resistance and reductions in Q _A and K _f . ^, Chronic administration of Ang II for 8 weeks resulted in systemic hypertension, lowered single-kidney GFR and, with the exception of K _f , elicited similar glomerular hemodynamic changes to those observed after acute infusion in both normal and uninephrectomized rats. The importance of the influence of endogenous Ang II on glomerular hemodynamics in remnant kidneys has been revealed by studies with pharmacologic inhibitors of the renin-angiotensin-aldosterone system (RAAS). Chronic treatment of 5/6 nephrectomized rats with either an angiotensin-converting enzyme (ACE) inhibitor ^, or Ang II (subtype 1) receptor blocker (ARB) ^, results in normalization of P _GC through a reduction in systemic BP and dilation of both afferent and efferent arterioles. SNGFR, however, remains elevated due to an increase in K _f . Furthermore, acute infusion of an ACE inhibitor or saralasin, a peptide analog receptor antagonist of Ang II, was found to normalize P _GC in 5/6 nephrectomized rats through efferent arteriolar dilation, without affecting the mean arterial pressure (MAP). ^, It is unclear why these findings could not be confirmed with the ARB losartan.

These effects of RAAS inhibition imply that there is increased local activity of endogenous Ang II, yet plasma renin levels show only a transient increase following 5/6 nephrectomy. ^, This suggests differential regulation of the systemic versus intrarenal RAAS and that Ang II is formed locally. Detailed studies have identified that all components of the RAAS are expressed in the kidney. Renin mRNA and protein levels are both increased in glomeruli adjacent to the infarction scar in 5/6 nephrectomized rats. Furthermore, kidney renin mRNA levels are increased at days 3 and 7 after kidney mass ablation by infarction but not when a kidney mass is excised surgically, suggesting that kidney infarction activates the RAAS by creating a margin of ischemic tissue around the organizing infarct and explaining the greater severity of hypertension and glomerulosclerosis associated with the infarction model. ^,

Detailed studies of intrarenal Ang II levels following 5/6 nephrectomy achieved by infarction have confirmed these findings by showing higher Ang II levels in the periinfarct portion of the kidney than the intact portion at all time points. On the other hand, the studies also showed that the rise in intrarenal Ang II levels following 5/6 nephrectomy was transient. Whereas Ang II levels in the periinfarct portion were elevated compared with sham-operated controls at 2 weeks after surgery, they were not statistically different at 5 or 7 weeks. In the intact portion of the remnant kidney, Ang II levels were similar to controls at 2 and 5 weeks and were lower at 7 weeks. Sustained increases in intrarenal Ang II levels are therefore not required to maintain hypertension and progressive kidney injury characteristic of this model. Nevertheless, subsequent studies have shown that the kidney-protective effects of ACE inhibitor and ARB treatment are associated with a reduction in intrarenal Ang II levels in both the periinfarct and intact portions of the remnant kidney. In contrast, treatment with the dihydropyridine calcium antagonist, nifedipine, did not reduce proteinuria, despite lowering BP to the same levels as the RAAS antagonists and was associated with an increase in intrarenal Ang II. Thus intrarenal Ang II appears to play a central role in the pathogenesis of hypertension and kidney injury in this model, even in the absence of sustained increases in Ang II levels. Further research is required to explain these findings fully. It could be argued that apparently normal intrarenal Ang II levels are inappropriately high in the context of hypertension and ECF volume expansion seen in these animals or that the total intrarenal Ang II levels measured may have failed to detect important local elevations of Ang II. Ongoing research has elucidated several novel aspects of the intrarenal RAAS, including regulation by several other factors like prorenin receptor, prostaglandin E2, and Wnt/β-catenin (positive regulators), as well as Klotho, vitamin D receptor, and liver X receptor (negative regulators).

Attention has focused on the potential role of aldosterone in progressive kidney injury. In addition to evidence that aldosterone may exert profibrotic effects in the kidney (see later), experiments have suggested that it may also have important glomerular hemodynamic effects. Previous observations that the deoxycorticosterone (DOCA)–salt model of hypertension is associated with glomerular capillary hypertension prompted detailed studies of microperfused rabbit afferent and efferent arterioles; these found dose-dependent constriction of both arterioles in response to nanomolar concentrations of aldosterone, with greater sensitivity observed in efferent arterioles. These effects were not inhibited by spironolactone and were still present with albumin-bound aldosterone, indicating that they may be mediated by specific membrane receptors rather than the intracellular receptors responsible for most of the actions of aldosterone. Interestingly, aldosterone may also counteract rabbit-afferent arteriolar vasoconstriction via a nitric oxide (NO)–dependent pathway, an action that would also be expected to increase P _GC . ^,

Endothelins (ETs) are potent vasoconstrictor peptides that act via at least two receptor subtypes, ET _A and ET _B . ET receptors have been identified throughout the body and are most abundant in the lungs and kidneys. ET _A receptors are primarily located on vascular smooth muscle cells and mediate vasoconstriction, as well as cellular proliferation, inflammation, and podocyte damage. ET _B receptors are expressed on vascular endothelial and kidney epithelial cells and appear to play a role as clearance receptors, as well as mediating endothelium-dependent vasodilation via NO and prostacyclin. Kidney production of endothelins is increased after 5/6 nephrectomy, raising the possibility that they may also contribute to the observed glomerular hemodynamic adaptations. ^, Acute and chronic infusion of endothelins elicits dose-dependent reductions in RPF and GFR in normal rats.

Observations regarding the relative effects of endothelins on afferent and efferent arterioles are to some extent contradictory, possibly reflecting different experimental conditions. Despite some differences, most studies in intact animals have reported greater increases in efferent than afferent arteriolar resistance, resulting in an increase in P _GC . The ultrafiltration coefficient (K _f ) was significantly reduced, and thus the SNGFR was unchanged or decreased. On the other hand, observations in microperfused arterioles have found that endothelins cause greater constriction of the afferent than efferent arteriole. Studies with selective ET _A and ET _B receptor antagonists and ET _B receptor knockout mice were somewhat contradictory, suggesting a complex interaction between ET _A and ET _B receptors in determining the response. ^, Comparative studies in rat juxtamedullary nephron preparations have shown that endothelin is a more potent vasoconstrictor of both afferent and efferent arterioles than Ang II, arginine vasopressin, norepinephrine, sphingosine-1-phosphate, and adenosine triphosphate (ATP). Using data from studies of endothelin infusion experiments in healthy human volunteers, detailed mathematical modeling identified the strongest effect to be afferent arteriolar vasoconstriction via ET _A and weaker vasoconstriction of the efferent arteriole. Systemic vascular resistance and renal vascular resistance both increased. In short-term studies of human subjects with CKD, endothelin receptor antagonists increased kidney blood flow but had little effect on the GFR due to a decrease in filtration fraction, observations consistent with a greater vasodilatory effect on the efferent arteriole. ^, Interestingly these observations were made in subjects already receiving treatment with an ACE inhibitor or ARB. The potential interaction between endothelins and other vasoactive molecules is further illustrated by observations that chronic infusion of Ang II results in increased production of endothelins and that ET-1 transgenic mice are not hypertensive but show induction of inducible nitric oxide synthase (iNOS), resulting in increased NO production as a probable counterregulatory mechanism to maintain normal BP. Furthermore, some of the glomerular hemodynamic effects of ETs appear to be modulated by prostaglandins. Detailed micropuncture studies to elucidate the role of endothelins in remnant kidney hemodynamics have not yet been published. These studies should be facilitated by the ongoing development of specific ET _A and ET _B receptor antagonists.

Atrial natriuretic peptide (ANP) and other structurally related NPs mediate, in large part, the functional adaptations in tubular sodium reabsorption that maintain sodium excretion in 5/6 nephrectomized rats but also exert important hemodynamic effects. Circulating ANP levels are elevated in 5/6 nephrectomized rats, and acute administration of a NP antagonist elicited profound decreases in the GFR and RPF in 5/6 nephrectomized rats on a high-salt (but not low-salt) diet, indicating that NPs play an important role in the observed hemodynamic responses to 5/6 nephrectomy.

In another study, brain natriuretic peptide (BNP) levels were found to be elevated after 3/4 nephrectomy in the absence of cardiac dysfunction or upregulation of myocardial BNP gene expression. Further insights into the kidney hemodynamic effects of NP have been gained from observations in normal rats infused with a synthetic ANP. Whole-kidney and single-nephron GFR increased by approximately 20% due entirely to a rise in P _GC , resulting from significant afferent arteriolar dilation and efferent arteriolar constriction. In the previous experiments, some residual elevation in remnant kidney GFR appeared to persist, even after the NP system was suppressed by sodium restriction or a NP receptor antagonist, suggesting that factors other than NP contribute to glomerular hyperfiltration following kidney mass ablation. The potential interaction between NPs and other vasoactive molecules is illustrated by the observation that ANP infusion in normal rats induces an increase in kidney nitric oxide synthase activity. In a human study pro-ANP and pro-BNP concentrations were observed to increase threefold at 24 hours after unilateral total or partial nephrectomy but returned to baseline after 5 days. Pro-ANP and pro-BNP concentration were elevated at 12 months after unilateral nephrectomy but not partial nephrectomy.

Eicosanoids, another family of potent vasoactive molecules present in abundance in the kidney, may also play a role in mediating glomerular hyperfiltration. Urinary excretion per nephron of both vasodilator and vasoconstrictor prostaglandins (PGs) is increased in rats and rabbits after kidney mass ablation. Infusion of PGE2, PGI2, or 6-keto-PGE1 into the renal artery elicits significant renal vasodilation. Whereas acute inhibition of prostaglandin synthesis by infusion of the cyclooxygenase (COX) inhibitor indomethacin has no effect on GFR or glomerular hemodynamics in normal rats, indomethacin lowers both the SNGFR and Q _A after 3/4 or 5/6 nephrectomy. ^, On the other hand, chronic treatment with a selective COX-2 inhibitor attenuates the systemic and glomerular hypertension observed in 5/6 nephrectomized rats but has no effect on the GFR.

The relative effects of prostaglandin synthesis inhibitors on afferent and efferent arterioles may vary with time post-nephrectomy. Afferent arteriolar constriction was the predominant finding reported at 24 hours post-surgery, whereas constriction of both afferent and efferent arterioles was observed at 3 to 4 weeks. ^, Some contribution of thromboxanes to glomerular hemodynamic adjustments after 5/6 nephrectomy in rats is suggested by the increase in the GFR seen after acute infusion of a selective thromboxane synthesis inhibitor. Thus different eicosanoids appear to exert opposite effects, but the general impression is that the combined effects of vasodilator prostaglandins outweigh those of the vasoconstrictors. This interaction is illustrated by the observation that perfusion of isolated glomeruli with bradykinin resulted in vasodilation of the efferent arteriole that was completely blocked by indomethacin but that this blockade was reversed by a specific antagonist of 20-hydroxyeicosatetraenoic acid (20-HETE), a vasoconstrictor eicosanoid, indicating that the glomerulus produces both vasodilator and vasoconstrictor eicosanoids. Thromboxane A2 (TXA2) may play a role in the development of hypertension after nephron loss. Plasma concentrations of the TXA2 metabolite, thromboxane B2, were elevated in rats after 5/6 nephrectomy and TXA2 receptor expression was increased in aortic rings.

The extremely short half-life of nitric oxide (NO) precludes direct measurement of nitric oxide levels or the administration of exogenous nitric oxide in experimental models. The actions of nitric oxide have thus been inferred from experiments with inhibitors of nitric oxide synthase (NOS). Intravenous infusion of NOS inhibitors results in systemic and renal vasoconstriction, as well as a reduction in the GFR, in normal rats. ^, Thus nitric oxide appears to exert a tonic effect on the physiologic maintenance of systemic BP and renal perfusion under resting conditions. It is unclear, however, whether nitric oxide plays a specific role in the adaptive hemodynamic changes that follow kidney mass ablation. Indeed, renal expression of NOS and renal nitric oxide generation is reduced in 5/6 nephrectomized rats, whereas systemic production of nitric oxide is increased. ^, MAP and renal vascular resistance increase, whereas renal blood flow (RBF) and the GFR decrease to a similar extent after acute infusion of an endothelial NOS (eNOS) inhibitor, NG-monomethyl- l -arginine ( l -NMMA), irrespective of whether given to normal rats or 3 to 4 weeks after unilateral or 5/6 nephrectomy. Chronic NOS inhibition with nitro- l -arginine methyl ester ( l -NAME) produces elevations in systemic BP and P _GC in 5/6 nephrectomized rats without affecting the GFR, whereas chronic treatment with aminoguanidine, an inhibitor of inducible NOS, has no effect on the GFR, RPF, or P _GC .

Similarly, greater increases in BP and proteinuria were observed after 5/6 nephrectomy in eNOS knockout versus wild-type mice. On the other hand, renal NOS expression and activity are increased early after unilateral nephrectomy, and pretreatment of rats with a subpressor dose of l -NAME prevents the early increase in RBF and decrease in renal vascular resistance usually observed after unilateral nephrectomy. ^, It therefore appears that nitric oxide plays a role in early hemodynamic adaptations to nephron loss, resulting in an increase in RBF, but in the longer term, nitric oxide retains a tonic influence on systemic and renal hemodynamics without being a specific determinant of the adaptive changes in glomerular hemodynamics.

Bradykinin is a potent vasodilatory peptide that is elevated in the remnant kidney and may therefore contribute to hemodynamic adaptations after nephron loss. Acute and chronic infusion of bradykinin result in increased RPF but have no effect on the GFR. ^, Micropuncture studies in intact animals are lacking, but studies of isolated perfused afferent arterioles have shown that bradykinin induces a biphasic response with vasodilation at low concentrations and vasoconstriction at higher concentrations. Both effects appear to be mediated by products of COX. Similar experiments with efferent arterioles have found dose-dependent vasodilation (no biphasic response) that was dependent on cytochrome P450 metabolites but independent of COX products or nitric oxide. When glomeruli were perfused with bradykinin, vasodilation of efferent arterioles was again observed but inhibited by a COX inhibitor, indicating that bradykinin induces glomerular production of COX metabolites (prostaglandins) that also contribute to efferent arteriolar dilation. Further studies are required to elucidate the role of bradykinin after nephron loss.

Urotensin II is the most potent vasoconstrictor identified to date, but its actions appear to vary in different vascular territories and, in some vessels, it may even produce vasodilation. Infusion of exogenous urotensin II has been reported to increase or decrease the GFR in normal rodents in different experiments (see Chapter 11). The vasoconstrictor effects of urotensin II may be mediated, at least in part, by upregulation of renal renin and aldosterone synthase gene expression. Urotensin II is produced in the kidney, and urotensin receptors have been localized to glomerular arterioles. Increased renal expression of mRNA for urotensin-related protein (which also binds to the urotensin receptor) and urotensin receptor has been reported in rats after 5/6 nephrectomy. Further experiments have observed increased immunostaining for urotensin II in tubules and glomeruli after 5/6 nephrectomy that increased as the kidney damage progressed. Increased urotensin receptor expression was also observed at more advanced stages. Treatment with a urotensin receptor antagonist delayed the onset of hypertension and albuminuria but did not attenuate glomerulosclerosis after 13 weeks. The role of urotensin II in the hemodynamic adaptations that follow nephron loss has yet to be investigated.

After extensive kidney mass ablation, there is a marked readjustment of the autoregulatory mechanisms that control RPF and the GFR. ^, The role of myogenic mechanisms is uncertain, but detailed studies of afferent arteriolar myogenic responses have suggested that their primary role is to protect the glomerulus from elevations in SBP. This notion has been supported by animal experiments that have reported that rat strains with impaired myogenic response show greater transmission of perfusion pressure to glomerular capillaries and are more susceptible to glomerular injury than strains with an intact myogenic response. The tubuloglomerular feedback system is reset after kidney mass ablation to permit and sustain the elevations in SNGFR and P _GC described previously. ^, Studies have indicated that connecting tubule glomerular feedback mediated by the epithelial sodium channel (ENaC) plays a key role in this process after uninephrectomy. Resetting appears to occur as early as 20 minutes after unilateral nephrectomy, in proportion to the extent of kidney ablation. The adjustments observed after uninephrectomy are of lesser magnitudes than those seen after 5/6 nephrectomy. The kidney-protective effects of sodium-glucose cotransporter 2 (SGLT2) inhibitors have been attributed at least in part to their natriuretic effect increasing tubuloglomerular feedback to reduce glomerular hyperfiltration. Treatment with the SGLT2 inhibitor, empagliflozin, after 5/6 nephrectomy resulted in an increase in urinary adenosine excretion, a marker of tubuloglomerular feedback activity, that was inversely related to the severity of tubulointerstitial fibrosis.

As is readily appreciated from the previous discussion, the adjustments in glomerular hemodynamics seen after kidney mass ablation represent the net effect of several endogenous vasoactive factors. NPs and vasodilator prostaglandins dilate the preglomerular vessels, whereas bradykinin dilates both afferent and efferent arterioles. On the other hand, Ang II, vasoconstrictor prostaglandins, and possibly endothelins constrict both afferent and efferent arterioles, with a greater effect on the latter. A net fall in preglomerular vascular resistance is observed, whereas efferent arteriolar resistance decreases to a lesser extent. Together with greater transmission of the raised systemic BP to the glomerular capillary network, these alterations in microvascular resistances result in the observed elevations in Q _A , P _GC , ΔP, and SNGFR ( Table 50.1 ). The importance of multiple vasoactive factors is illustrated by the observation that treatment of 5/6 nephrectomized rats with omapatrilat, an inhibitor of both ACE and neutral endopeptidase that results in reduced Ang II production, as well as increased NPs and bradykinin levels, lowers P _GC more than ACE inhibition alone.

The complexity of factors involved is further illustrated by observations that other molecules involved in the modulation of progressive kidney injury may exert hemodynamic effects by influencing the mediators discussed previously. Acute infusion of hepatocyte growth factor has been shown to induce a decline in BP and GFR, an effect that is mediated by a short-term increase in ET-1 production. In isolated perfused preparations, platelet-activating factor at picomolar concentrations has been shown to induce glomerular production of NO, resulting in dilation of preconstricted efferent arterioles, whereas at nanomolar concentrations, platelet-activating factor constricts efferent arterioles through local release of COX metabolites. The potential role of other recently identified vasoactive molecules such as urotensin II remains to be elucidated.

Among the earliest responses to unilateral nephrectomy are biochemical changes that precede cell growth. Increased incorporation of choline, a precursor of cell membrane phospholipid, has been detected as early as 5 minutes, ^, and increased choline kinase activity has been observed at 2 hours after nephrectomy. Activity of ornithine decarboxylase, the enzyme catalyzing the first step of polyamine synthesis, is elevated at 45 to 120 minutes, and polyamine levels peak at 1 to 2 days post-nephrectomy. Early alterations in mRNA metabolism have also been observed. Although there are no changes in the half-life or cytoplasmic distribution of mRNA, a near-25% increase in the fraction of newly synthesized poly(A)-deficient mRNA occurs within 1 hour of uninephrectomy, and total RNA synthesis in the kidney increases by 25% to 100% relative to that in the liver. Ribosomal RNA synthesis is increased by 40% to 50% at 6 hours. The rate of protein synthesis is increased at 2 hours and is nearly doubled at 3 hours. Data on cyclic nucleotide levels, which are thought to affect cell growth and proliferation, are conflicting. Some studies have reported elevated levels of cyclic guanosine monophosphate (cGMP) in the remaining kidney as early as 10 minutes after surgery, whereas others have found no consistent changes in cyclic adenosine monophosphate (cAMP) or cGMP levels. ^, Genome-wide analysis of gene expression using cDNA microarrays in remaining rat kidneys up to 72 hours after uninephrectomy has revealed the dominant response to be suppression of the genes responsible for inhibition of growth and apoptosis.

Early biochemical changes are followed by a period of rapid growth. DNA synthesis is increased at 24 hours, and increased numbers of mitotic figures are evident at 28 to 36 hours. Both reach a maximum 5- to 10-fold increase at 40 to 72 hours. In rats, kidney weight is increased at 48 to 72 hours after uninephrectomy and increases by 30% to 40% at 2 to 3 weeks ( Fig. 50.2 ). ^, The nephron number is fixed shortly before birth in most species, so this gain in kidney weight is attributable to increased nephron size. Growth is thought to occur largely through cell hypertrophy, accounting for 80% of the increase in kidney mass seen in adult rats and, to a lesser extent, through hyperplasia. Kidney mass continues to rise for 1 to 2 months until a 40% to 50% increase is achieved. The degree of compensatory growth is a function of the extent of kidney ablation. Uninephrectomy has been shown to provoke an 81% increase of residual kidney mass at 4 weeks compared with an increase of 168% after 70% kidney ablation. Normal controls gained 31% in kidney weight over the same period.

Rate of compensatory renal growth after unilateral nephrectomy (circles) and ureter ligation ( squares) .

From Dicker SE, Shirley DG. Compensatory hypertrophy of the contralateral kidney after unilateral ureteral ligation. J Physiol (Lond). 1972;220:199−210.

Age diminishes kidney hypertrophic responses. After uninephrectomy, greater increases in kidney weight and more extensive hyperplasia were observed in 5- versus 55-day-old rats, and aging rats exhibited gains in kidney weight of only one-third to three-quarters of those seen in younger controls. ^,

In humans, assessment of kidney hypertrophy after nephrectomy is dependent on radiologic studies. Ultrasound studies have reported increases of 19% to 100% in kidney volume and, in computed tomography (CT) studies, have found an increase of 30% to 53% in kidney cross-sectional area after nephrectomy. ^, Contrast-enhanced CT has been used to measure kidney parenchymal volume post unilateral nephrectomy. One study reported increases of 12.1% and 8.9% at 1 week and 6 months, respectively. The degree of hypertrophy correlated positively with the function of the kidney removed and negatively with patient age. One relatively large study of living kidney donors observed a 27.6% ± 9.7% increase in the remaining kidney volume at 6 months post donor nephrectomy, and a further study has reported an increase in kidney cortical volume of 27% at a median of 0.8 years post nephrectomy that increased to 35% after a median of 6.1 years. Other studies using CT scans have reported increases in kidney volume of 16.5 to 18.5% at 1 year after nephrectomy for renal cell carcinoma and 22.4% ± 23.2% at 6 to 12 months after donor nephrectomy. In a detailed study that used magnetic resonance imaging (MRI) to measure kidney volume, an increase was observed from 198 ± 87 mL preoperatively to 329 ± 175 mL at 3 months post nephrectomy for renal cell carcinoma. The change in kidney volume correlated positively with changes in RBF at 1 week post nephrectomy (r = 0.6). Differences among these studies are likely attributable to the relatively small number of subjects enrolled, wide variation in the time intervals between nephrectomy and assessment of kidney size, and differing indications for nephrectomy.

The principal morphometric change observed in glomeruli after uninephrectomy is an increase in volume. Glomerular enlargement appears to parallel whole-kidney growth and has been detected as early as 4 days after surgery. The degree of enlargement of superficial and juxtamedullary glomeruli is similar. Proportionally similar increases in number and size of all cell types occur, with preservation of the relative volumes of different glomerular cells. There is consensus that glomerular capillaries increase in length and number (i.e., more branching), but most studies have shown that diameter or cross-sectional surface area of the glomerular capillaries remains constant or increases only minimally. ^, Transplantation of hypertrophied kidneys into uninephrectomized rats has demonstrated regression of glomerular hypertrophy within 3 weeks, yet the increase in capillary length was maintained.

Glomerular hypertrophy, as evidenced by elevated RNA/DNA and protein/DNA ratios, as well as by increased glomerular volume (V _G ) on electron microscopy, has been detected at 2 days after 5/6 nephrectomy in rats. The initial increase in V _G was due almost entirely to increases in visceral epithelial cell volume, whereas at 14 days the increase in V _G was largely accounted for by mesangial matrix expansion. Although several studies have reported glomerular capillary lengthening after 5/6 nephrectomy, few have detected any increase in cross-sectional area or diameter of the glomerular capillaries. These observations should, however, be considered in the light of important technical considerations. In vitro perfusion of isolated glomeruli has demonstrated that V _G increases as perfusion pressure is raised through physiologic and pathophysiologic ranges. Moreover, glomerular capillary “compliance” in these studies was a function of the baseline V _G , and glomeruli obtained from remnant kidneys post 5/6 nephrectomy had higher compliance than those from control animals. These findings have two important implications. First, although glomerular pressures are only minimally elevated after uninephrectomy, the glomerular capillary hypertension associated with more extensive kidney ablation is likely to contribute significantly to the increase in V _G . Second, estimates of V _G in tissues that have not been perfusion-fixed at the appropriate BP should be interpreted with caution. Direct comparison of V _G in perfusion-fixed versus immersion-fixed kidneys from the same rats yielded estimates of V _G in immersion-fixed samples that were 61% lower than those from perfusion-fixed kidneys.

The notion that hypertrophy after uninephrectomy is stimulated by the need for the remaining kidney to excrete larger amounts of metabolic waste products, necessitating more excretory “work,” was proposed by Sacerdotti in 1896. Subsequently, it became apparent that urea excretion is largely a function of glomerular filtration, whereas the main energy-requiring function of the renal tubules is reabsorption of filtered electrolytes (principally sodium), and water. The hypothesis was therefore modified to view hypertrophy as a response to the increased demand for water and solute reclamation imposed by an increased SNGFR (solute load hypothesis). Several lines of evidence have supported the concepts underlying the solute load hypothesis. After uninephrectomy, RBF increased by 8% in the remaining kidney and preceded hypertrophy, and treatment with a subpressor dose of the NOS inhibitor l -NAME prevented the rise in RBF and substantially attenuated increases in kidney weight, as well as glomerular and proximal tubule areas, at 7 days postnephrectomy. In the remnant kidney, proximal tubule sodium absorption increased in parallel with GFR (glomerulotubular balance), and tubules continued to display enhanced fluid reabsorption in vitro, implying that the adaptive changes were intrinsic to the tubular epithelial cells (TECs). ^, In chronic glomerulonephritis, a lesion characterized by marked heterogeneity in the SNGFR, there was preservation of the SNGFR to proximal fluid reabsorption ratio and a close correlation between glomerular and proximal tubule hypertrophy. Moreover, sustained increases in the GFR in the absence of kidney mass ablation result in kidney hypertrophy in some conditions including pregnancy (in some but not all studies) and diabetes mellitus.

On the other hand, experimental maneuvers dissociating kidney solute load from hypertrophy appear to contradict the solute load hypothesis. Total diversion of urine from one kidney into the peritoneum by ureteroperitoneostomy was associated with an increase in the GFR in the contralateral kidney of similar magnitude to that seen after uninephrectomy, but no increase in kidney mass or mitotic activity. In another example, potassium depletion resulted in kidney hypertrophy without any increase in GFR. Moreover, the findings that some of the early biochemical changes associated with hypertrophy precede increases in glomerular filtration or sodium reabsorption argue against a causal association of hypertrophy and increased solute load. Despite these conflicting data, there is nevertheless considerable evidence of an association between SNGFR and proximal tubule hypertrophy that may play a role in stimulating kidney growth in the remnant kidney.

Failure of the solute load hypothesis to explain all the experimental data has led others to propose instead that the primary stimulus for kidney hypertrophy is a change in kidney mass and that kidney growth is under the control of specific growth and/or inhibitory factors. Evidence in support of this theory was derived from three types of experiments.

In the first, a stable connection was established between the extracellular space and microcirculation of two animals (parabiosis), and the effects of kidney mass ablation in one animal were assessed in the intact kidneys of its partner. Despite some inconsistencies due to variations in methodology, these experiments generally found that uninephrectomy in one animal resulted in hypertrophy of the contralateral kidney and, to a lesser extent, of both kidneys of the parabiotic partner. Bilateral nephrectomy in one partner or a triple nephrectomy produced incremental degrees of hypertrophy in the remaining kidney(s). Furthermore, the hypertrophy was rapidly reversed following cessation of cross circulation.

A second strategy was to inject serum or plasma from uninephrectomized animals into intact subjects and then assess kidney hypertrophy by radiolabeled thymidine uptake or mitotic count. Although studies using single small intraperitoneal or subcutaneous doses were negative, the administration of repeated large doses by an intraperitoneal or intravenous route elicited kidney hypertrophy in most subjects studied.

The data that most consistently supported the existence of a renotropic factor were derived from in vitro experiments in which kidney tissues were incubated in the presence or absence of plasma or serum from rats subjected to kidney mass ablation. Evidence for hypertrophy was generally assessed by the incorporation of radiolabeled thymidine or uridine into DNA or RNA, respectively. In general, these experiments showed increased uptake of radiolabeled nucleotides after incubation with serum from uninephrectomized animals. This effect appeared to be organ but not species specific. That a tissue factor produced by the kidneys and upregulated after nephrectomy may be required for the activity of a circulating renotropin was suggested by experiments in which kidney extract from rats taken 20 hours after uninephrectomy, in the presence of normal rat serum, was found to stimulate ³ H-thymidine incorporation in normal kidney cortex. However, addition of the same extract in the absence of the serum tended to depress ³ H-thymidine uptake. Serum taken from bilaterally nephrectomized animals lacked renotropic effects, but these were restored after dialysis of the serum, suggesting the presence of renotropin inhibitory factors that accumulated in the absence of kidney function. Although the specific identity of “renotropin” has remained elusive, several lines of evidence suggested that it was a small protein. Retention of activity after ultrafiltration, dialysis, and removal of albumin from serum implied that renotropin was a molecule of 12 to 25 kDa in size, with no significant binding to albumin. ^,

Several hypotheses were advanced to reconcile the previously observed effects and operation of a putative renotropic system. ^, The following were variously proposed: 1. Renotropin is a circulating substance normally catabolized or excreted by the kidneys; 2. kidney growth is regulated by a specific renotropin-producing tissue inhibited by a factor produced by normal kidneys; and 3. kidney growth is tonically inhibited by a substance produced by normal kidneys, a decrease in the levels of which induce an enzyme in the kidney cortex that cleaves a circulating precursor of renotropin to produce the active molecule.

Several of the major endocrine systems influence kidney growth, but each lacks selective effects on the kidney. There is little evidence that any of these systems represents the specific mediators of compensatory kidney hypertrophy. Whereas early experiments suggested that hypophysectomy inhibits compensatory hypertrophy after uninephrectomy, later studies that controlled for the reduction in kidney mass that usually accompanies hypopituitarism found a degree of hypertrophy comparable with that seen in normal rats.

Nevertheless, specific renotropic activity has been identified in a subfraction of ovine pituitary extract associated with a lutropin-like substance. ^, Uninephrectomy is accompanied by a transient increase in the pulsatile release of growth hormone in male but not female rats, suggesting a role for this hormone in the early phase of hypertrophy in males. When the increase in growth hormone was prevented by the administration of an antagonist to growth hormone–releasing factor or the effects of growth hormone were blocked by a growth hormone receptor blocker, kidney hypertrophy was significantly attenuated. ^, Adrenal hormones appear to play only a small role in kidney hypertrophy. Adrenalectomy does not inhibit compensatory growth after uninephrectomy. Whereas kidney weight relative to body weight is reduced in hypothyroidism and increased by excess thyroid hormone, compensatory hypertrophy still occurs in thyroidectomized rats. Progesterone and estradiol in excess or ovariectomy have little effect on kidney weight, but testosterone appears to play a role, as evidenced by a fall in kidney and body weight after orchidectomy and an increase in kidney weight with excess testosterone. ^, Whereas orchidectomy does not inhibit hypertrophy after uninephrectomy, exogenous testosterone does increase the degree of hypertrophy observed in some but not all studies. ^,

Of the numerous growth factors and their receptors that have been localized in the kidney, at least four are associated with kidney hypertrophy. ^, Several lines of evidence suggest a role for insulin-like growth factor-1 (IGF-1). Kidney tissue IGF-1 levels were elevated at 1 to 5 days after uninephrectomy and started to decline within days in some ^, but not all studies. In one study, the level of kidney IGF-1 expression correlated significantly with the extent of kidney mass ablation. On the other hand, Shohat and associates have found an increase in serum IGF-1 levels only at 10 days post nephrectomy, which was still present on day 60. That IGF-1 may be induced independently of GH in the setting of kidney hypertrophy is illustrated by preservation of the increase in kidney IGF-1 in hypophysectomized and GH-deficient rats.

Other molecules related to IGF function are also upregulated. Kidney IGF-1 receptor gene expression was increased twofold to fourfold in female rats after uninephrectomy ; IGF-1 binding protein mRNA was upregulated in the remnant kidney at 2 weeks after 5/6 nephrectomy ; and analysis of the genome-wide transcriptional response to unilateral nephrectomy identified IGF-2 binding protein as one of the activated genes. Further evidence has suggested that IGF-1 may in turn promote production of vascular endothelial growth factor (VEGF), implying that VEGF may be a downstream mediator of IGF-1 effects, at least in the pathogenesis of diabetic retinopathy. That VEGF is important for compensatory kidney hypertrophy has been confirmed by the observation that treatment of mice with VEGF antibodies after uninephrectomy completely prevents glomerular hypertrophy and inhibits kidney growth at 7 days. Epidermal growth factor (EGF) in the remaining kidney is increased on day 1 in mice and by day 5 in rats. In addition, EGF has been shown to induce IGF-I mRNA production in collecting duct cells in vitro, suggesting the existence of a local paracrine system. Increased mRNA levels for both hepatocyte growth factor and its receptor, c-met, have been demonstrated in the remaining kidney as early as 6 hours after uninephrectomy. ^, In another study, the rise in hepatocyte growth factor message was found to be nonspecific, occurring in both liver and kidney and also in sham-operated rats, whereas the increase in mRNA for c-met was specific for the outer kidney medulla.

Despite these associations, the timing of the changes in growth factor levels remains unclear. Whereas some investigators have reported early increases, ^, several others reported changes only at time points when significant hypertrophy is already present, thus failing to provide convincing evidence that they represent the proximal effectors in a renotropic system. ^,

Mesangial cells play a central role in glomerular function, modulating glomerular capillary blood flow and ultrafiltration surface area. In addition, mesangial cells are both a source of and target for vasoactive molecules, growth factors, cytokines, and extracellular matrix proteins. Studies have suggested that they also play a major role in compensatory kidney hypertrophy. In vitro experiments have found that when mesangial cells from a remaining kidney after uninephrectomy are cultured with serum obtained from rats after uninephrectomy, their conditioned medium induces hypertrophy in tubule cells. Uninephrectomy induces significant transient proliferation of mesangial cells, reaching a peak at 24 hours and ceasing within 72 hours. This proliferation occurs in an environment of increased circulating and kidney growth factors and cytokines, such as growth hormone, IGF-1, and interleukin-10 (IL-10), as well as reduced levels of antiproliferative factors such as transforming growth factor-β (TGF)-β and ANP. A reduction in mesangial cell proliferation occurs in parallel with the onset of tubule cell hypertrophy. IL-10 and TFG-β have been identified as the major mediators of the mesangial regulation of tubule cell hypertrophy. Mesangial cells are the main source of IL-10 in the kidney, where it acts as an autocrine growth factor and induces the expression of TFG-β. Mesangial cells are the only resident kidney cells known to produce and activate TFG-β, a process that is regulated by multiple factors, including Ang II, IGF-1, HGF, basic fibroblast growth factor (bFGF), tumor necrosis factor-α (TNF-α), EGF, and platelet-derived growth factor (PDGF), all of which are produced by mesangial cells. IL-10 expression starts to increase in the remaining kidney within hours of uninephrectomy, peaks at 24 hours and returns to normal within several days. In contrast, circulating and kidney TFG-β levels fall in the first 24 hours after nephrectomy and start to rise from 72 hours, reaching a peak at 1 week.

The importance of IL-10 was confirmed by experiments in which inhibition of IL-10 production resulted in lower TFG-β levels and reduced tubular hypertrophy, resulting in a 20% to 25% reduction in the weight of the remaining kidney. IL-10 has no direct effect on tubule cells, whereas TFG-β has been identified as an important mediator of tubule cell hypertrophy. The data presented are therefore consistent with the hypothesis that hyperfiltration after unilateral nephrectomy induces mesangial cell proliferation, as well as the production of IL-10 and other growth factors that induce expression and activation of TFG-β in mesangial cells. TFG-β in turn stimulates tubule cell hypertrophy ( Fig. 50.3 ). This goes a long way to unifying the components of the solute load and renotropin hypotheses into a single paradigm to explain the mechanisms of compensatory kidney hypertrophy.

Possible mechanisms involved in compensatory renal growth after unilateral nephrectomy.

Ang II, Angiotensin II; EGF, epidermal growth factor; IGF-1, insulin-like growth factor-1; IL-1, interleukin-1; TGF- β, transforming growth factor-β.

From Sinuani I, Beberashvili I, Averbukh Z, et al. Mesangial cells initiate compensatory tubular cell hypertrophy. Am J Nephrol. 2010;31:326−331.

Detailed investigation of cellular responses to kidney mass reduction has begun to elucidate some of the mechanisms involved in compensatory hypertrophy at the cellular level. Kidney hypertrophy is achieved by modulation of the cell cycle, which becomes arrested in the late G1 phase and therefore does not progress to the S phase, resulting in cell hypertrophy instead of hyperplasia. This is achieved through activation of cyclin-dependent kinase (CDK) 4−cyclin D without subsequent engagement of CDK 2−cyclin E, a process thought to be regulated by TGF-β and CDK inhibitor proteins p21 ^Waf1 , p27 ^kip1 , and p57 ^kip2 . Activity of CDK 4−cyclin D complexes increases at 4, 7, and 10 days post nephrectomy, and CDK 2−cyclin E increases at days 2, 4, and 7 to 14, implying that p21 ^Waf1 , p27 ^kip1 , and p57 ^kip2 may play an important role in regulating tubule cell hypertrophy. The required increase in RNA and protein synthesis appears to be mediated at least in part by mammalian target of rapamycin (mTOR), a protein kinase that controls protein synthesis, as well as cell growth and metabolism.

In cells, mTOR exists in two distinct multiprotein complexes, mTORC1 and mTORC2. mTORC1 acts through multiple mediators to regulate protein synthesis and cell size. Two important downstream effectors of mTORC1 are 4E-binding protein 1 and protein kinase p70S6 kinase 1 (S6K1). Transcriptomic analysis after nephrectomy in mice has identified mTORC1 as the most significantly increased gene set at 24 hours, with elevations sustained at 48 and 72 hours. The importance of mTORC1 has been confirmed by the observation that pretreatment of rats with rapamycin, an inhibitor of mTORC1, inhibits kidney hypertrophy after uninephrectomy. Furthermore, experiments in S6K1 knockout mice have found inhibition of 60% to 70% of the hypertrophy observed after uninephrectomy, indicating that S6K1 plays a major role in compensatory hypertrophy.

Studies have also identified early upregulation in genes involved in cholesterol biosynthesis after unilateral nephrectomy. A detailed multiomics study has utilized transcriptomics and proteomics to identify the lipid-activated transcription factor, peroxisome proliferator-activated receptor α (PPARα), as a key mediator involved in hypertrophy of proximal tubule cells in mice after uninephrectomy. Furthermore, treatment with a PPARα activator, fenofibrate, significantly enhanced hypertrophy, whereas hypertrophy was reduced in mice with deletion of the gene for PPARα.

Several mechanisms have been proposed whereby glomerular hypertension and hyperperfusion may result in glomerular cell injury. Experiments in isolated perfused rat glomeruli have reported significant increases in glomerular volume, with increases in perfusion pressure over the normal and relevant abnormal range. These increases in wall tension and glomerular volume likely result in stretching of glomerular cells. Experimental evidence has suggested that such stretching may have adverse consequences for all three major cell types in the glomerulus. A mathematic model of blood flow and filtration in the glomerular capillary network after 5/6 nephrectomy has estimated that glomerular capillary wall strain is increased approximately threefold versus control, a magnitude sufficient to impact podocyte structure and function. Furthermore, advances in the study of cellular responses to mechanical stress raised the possibility that glomerular hyperperfusion may also promote the development of glomerulosclerosis through more subtle and complex pathways that induce profibrotic phenotypic alterations in glomerular cells.

The vascular endothelium serves a number of complex functions, including acting as a dynamic barrier to leukocytes and plasma proteins, secretion of vasoactive factors (prostacyclin, nitric oxide, and endothelin), conversion of Ang I to Ang II, and expression of cell adhesion molecules. It is also the first cellular structure in the kidney that encounters the mechanical forces imparted by glomerular hyperperfusion. After 5/6 nephrectomy, endothelial cells are activated or injured, resulting in detachment and exposure of the basement membrane ( Fig. 50.6 ). This in turn may induce platelet aggregation, deposition of fibrin, and intracapillary microthrombus formation. ^, It has been recognized for some time that segmental glomerulosclerosis is associated with focal obliteration of capillary loops and that interstitial fibrosis is associated with a loss of peritubular capillaries. Furthermore, it has been shown that this loss of capillaries in the remnant kidney is associated with a decrease in endothelial cell proliferation and reduced constitutive expression of VEGF by podocytes and renal tubule cells, as well as increased expression of the antiangiogenic factor, thrombospondin-1, by the kidney interstitium.

Because VEGF is an important endothelial cell angiogenic, survival, and trophic factor, these findings suggest that capillary loss may be caused in part by failure of recovery from hemodynamically mediated endothelial cell injury. Indeed, inhibition of endothelial and mesangial cell proliferation by treatment with everolimus after 5/6 nephrectomy resulted in increased proteinuria, glomerulosclerosis, and interstitial fibrosis associated with reduced glomerular expression of VEGF. Furthermore, short-term treatment of rats with VEGF ameliorated both glomerular and peritubular capillary loss after 5/6 nephrectomy. This preservation of capillaries was associated with a trend toward less glomerulosclerosis and significantly less interstitial deposition of type III collagen, as well as better preservation of kidney function. Further evidence of the importance of endothelial cells in the preservation of function after nephron loss has been provided by experiments in which bone marrow−derived endothelial progenitor cells (EPCs) were administered to mice after 5/6 nephrectomy. EPC-treated mice showed better preservation of kidney function, less proteinuria, and relatively preserved kidney structure, as well as decreased expression of proinflammatory molecules and restored levels of the angiogenic molecules VEGF, VEGF receptor-2, and thrombospondin 1. Long-term studies are required to evaluate further the potential benefits of improving renal angiogenesis in the setting of progressive kidney damage.

Endothelial cells bear numerous receptors that allow them to detect and respond to changes in mechanical forces. Thus exposure of endothelial cells to changes in shear stress, cyclic stretch, or pulsatile barostress that result from glomerular hyperperfusion may induce changes in the expression of genes involved in inflammation, cell-cycle control, apoptosis, thrombosis, and oxidative stress. The in vitro responses of endothelial cells to mechanical forces have largely been studied in the context of vascular remodeling and atherosclerosis, but it can readily be appreciated that similar responses may affect the development of inflammation and fibrosis in the remnant kidney. Of particular interest are observations that shear stress can stimulate endothelial expression of adhesion molecules and proinflammatory cytokines. It is clear therefore why biomechanical activation has emerged as an important paradigm in endothelial cell biology, but further studies focusing on glomerular endothelial responses to mechanical stress are required to elucidate the role of these mechanisms in progressive kidney damage.

Mesangial cells are closely associated with the capillaries in the glomerulus and are therefore also exposed to mechanical forces. Evidence from in vitro studies has indicated that mesangial cells respond to changes in these mechanical forces in ways that may promote inflammation and fibrosis. Subjecting mesangial cells to cyclical stretch or strain has been shown to induce the proliferation and synthesis of extracellular matrix constituents. ^, Cyclic stretch also activates the transcription factor nuclear factor-κβ (NF-κβ) and stimulates monocyte chemoattractant protein 1 (MCP-1), TGF-β, and its receptor, as well as connective tissue growth factor (CTGF) production. In cultured mesangial cells, cyclic stretch activates the RAAS and Ang II in turn may induce TGF-β synthesis.

In vitro studies have identified several signaling pathways responsible for stretch-induced signal transduction in mesangial cells. Cyclic stretch activates the serine-threonine kinase Akt through a mechanism requiring phosphatidylinositol-3-kinase and transactivation of EGF receptor and is necessary for the increased synthesis of collagen type 1A1 observed in the mesangial cells. Furthermore, Akt activation was observed in remnant kidney glomeruli, indicating that these mechanisms are also present in vivo. The actin cytoskeleton is an important transmitter of mechanical signals, and the GTPase, RhoA, a central regulator of the cytoskeleton, is activated by cyclic stretch. Stretch-induced activation of the mitogen-activated protein kinase Erk, linked to increased matrix production, is dependent on the activation of RhoA. Mesangial cells cultured at ambient pressures of 50 to 60 mm Hg (i.e., levels corresponding to glomerular capillary hypertension) also show enhanced synthesis and secretion of extracellular matrix when compared with cells grown at “normal” pressures of 40 to 50 mm Hg. Exposure of mesangial cells to barostress, achieved by culture under increased barometric pressure, stimulates the expression of cytokines including PDGF-B ³²⁷ and MCP-1. Transduction of mechanical forces by mesangial cells has been associated with tyrosine phosphorylation and protein kinase C–induced increases in S-6 kinase activity.

A growing body of evidence attests to the importance of podocyte injury in a variety of kidney diseases and in CKD progression. Podocytes display morphologic evidence of injury as early as 1 week after 5/6 nephrectomy (see Fig. 50.5 ) and 6 months after uninephrectomy. Increased numbers of podocytes have been observed in the urine of rats after 5/6 nephrectomy and in human CKD. In 5/6 nephrectomized rats, the number of podocytes in the urine correlated with the severity of proteinuria, as well as mean arterial BP, suggesting that podocyte loss may contribute to CKD progression. The importance of podocyte injury in CKD progression is further supported by the observation that amelioration of glomerular damage in 5/6 nephrectomized rats treated with 1,25-dihydroxyvitamin D3 is associated with preservation of podocyte number, as well as prevention of podocyte hypertrophy and injury. Detailed in vitro studies have shown that early podocyte injury is associated with dysregulation of calcium homeostasis and disruption of the actomyosin contractile apparatus. The Rho family of small guanosine triphosphatases plays a key role in this process. Calcium influx via transient receptor potential canonical type 5 (TRPC5) channels results in the activation of Rac1 and is associated with increased podocyte migration and proteinuria, whereas calcium influx via TRCP type 6 activates RhoA, resulting in the preservation of stress fibers, prevention of podocyte migration, and maintenance of the filtration barrier. However, subsequent experiments in TRCP6-deficient podocytes have found that TRCP6 is not required for mechanotransduction. Rather, mechanical stretch was found to induce ATP release from podocytes by a process dependent on cholesterol and podocin. ATP release in turn activated the purinergic channel P2X ₄ , resulting in an influx of calcium and increased disorganization of the actin cytoskeleton.

Thus purinergic channels appear to be key mechanotransducers in podocytes. Further research has identified the mechanosensing ion channel, Peizo 1, as an important mechanotransducer in podocytes. Piezo 1 was found to be upregulated in a rodent model of hypertensive nephropathy, and this was associated with increased expression of podocyte injury markers. Because podocytes are attached to the outer aspect of the glomerular basement membrane, it is reasonable to expect that they would be exposed to increased mechanical forces resulting from glomerular hypertension. Confirmation that podocytes respond to such physical forces has been derived from several in vitro experiments that examined podocyte responses to stretching. Activation of a voltage-sensitive potassium channels was observed in response to stretching of the podocyte cell membrane, and culture of podocytes under constant stretch-induced activation of the protein kinases Erk1/2 and JNK via the EGF receptor, as well as other changes in cell signaling. Mechanical stretch inhibited podocyte proliferation and, in common with TGF-β, reduced α ₃ β ₁ -integrin expression and podocyte adhesion in vitro. Exposure to cyclic stretching that mimics pulsatile strain within the glomerulus has been shown to cause reorganization of the actin cytoskeleton, upregulation of COX-2, and E prostanoid (EP)-4 receptor expression, as well as podocyte hypertrophy. Subsequent experiments using mice with podocyte-specific depletion or overexpression of EP4 receptors have found that prostaglandin E2 acting via EP4 receptors contributes to the development of proteinuria after 5/6 nephrectomy and therefore probably contributes to podocyte injury.

In another experiment, cyclic stretching of podocytes was associated with increased production of Ang II and TGF-β, as well as upregulation of angiotensin subtype 1 (AT1) receptors, resulting in increased Ang II−dependent apoptosis. Cyclic stretch of podocytes also resulted in a 50% reduction of nephrin (a key component of the slit diaphragm) mRNA and protein levels via an Ang II−dependent mechanism that was inhibited by the peroxisome proliferator-activated receptor-gamma (PPAR-γ) agonist, rosiglitazone, which prevented AT1 receptor upregulation.

The importance of podocyte loss in CKD progression has been demonstrated by experiments in which podocytes were depleted using diphtheria toxin. Loss of >30% of podocytes destabilized glomeruli and provoked progressive podocyte loss even after the initial injury eventuating in glomerulosclerosis. A similar pattern of podocyte loss was observed in the 5/6 nephrectomy model and was ameliorated by treatment with an ACE inhibitor and angiotensin receptor blocker. Taken together, these data suggest that stretch-induced podocyte injury is a further mechanism whereby glomerular hypertension contributes to glomerular injury.

Despite the lack of an obvious immune stimulus, an inflammatory cell infiltrate composed predominantly of macrophages and smaller numbers of lymphocytes is observed in remnant kidneys after 5/6 nephrectomy. Interestingly, similar observations have been reported in a rat model of spontaneous kidney agenesis associated with a 60.2% reduction in nephron endowment. As discussed earlier, it is possible that the glomerular hemodynamic adaptations to nephron loss may provoke an inflammatory cell response through the effects of mechanical forces on endothelial and mesangial cells. Thus upregulation of renal endothelial adhesion molecules may facilitate the egress of leukocytes from the circulation into the mesangium, where they may participate in further kidney injury. The recruited cellular infiltrate may constitute an abundant source of potent pleiotropic cytokine products, which in turn influence other infiltrating leukocytes, dendritic cells, and kidney cells, stimulating cell proliferation, elaboration of extracellular matrix components, and increased endothelial adhesiveness.

Evidence is emerging that these proposed mechanisms, based largely on in vitro observations, are indeed relevant in vivo. In the two-kidney, one-clip model of renovascular hypertension, upregulated expression of adhesion molecules and TGF-β, as well as cell infiltration, is observed only in the nonclipped kidney that is exposed to hypertensive perfusion pressure. ^, In the 5/6 nephrectomy model, coordinated upregulation of a variety of cell adhesion molecules, cytokines, and growth factors in association with macrophage infiltration has been observed at time points that precede the development of severe glomerulosclerosis. ^, Furthermore, the kidney protection afforded by an ACE inhibitor or ARB treatment in this model was associated with inhibition of cytokine upregulation and prevention of renal infiltration by macrophages. ^,

Infiltrating macrophages, although present in the glomeruli of remnant kidney, are chiefly distributed in the tubulointerstitial regions, ^, suggesting that they play a role in the development of the TIF that accompanies glomerulosclerosis. Further analysis of the cellular infiltrate has also identified mast cells near areas of TIF. It is possible that interstitial infiltrates are recruited as the result of tubulointerstitial cell activation by the downstream effects of cytokines released in the glomeruli. Alternatively, it has been proposed that excessive uptake of filtered proteins by tubule epithelial cells stimulates expression of cell adhesion and chemoattractant molecules that recruit macrophages and other monocytic cells to tubulointerstitial areas (see later for further discussion). The chemokine receptor CCR-1 has been shown to be important in interstitial but not glomerular recruitment of leukocytes. Treatment with a nonpeptide CCR-1 antagonist has been shown to reduce interstitial macrophage infiltration and ameliorate interstitial fibrosis in the UUO model, but data are still lacking in the 5/6 nephrectomy model. Furthermore, antagonism of MCP-1 signaling through gene therapy—induced production of a mutant form of MCP-1 by skeletal muscle resulted in reduced interstitial macrophage infiltration and amelioration of interstitial fibrosis in mice after UUO.

The identification of renal tubule cells expressing α-smooth muscle actin after 5/6 nephrectomy has raised the possibility that tubule cells may undergo transdifferentiation to a myofibroblast phenotype that contributes to interstitial fibrosis. Furthermore, the kidney protection observed with mycophenolate treatment in 5/6 nephrectomized rats is associated with reductions in interstitial myofibroblast infiltration and collagen type III deposition. There is, however, ongoing debate regarding the origin of interstitial myofibroblasts in CKD. Fate-mapping studies have indicated that mesenchymal cells called pericytes are the predominant source of myofibroblasts and that epithelial-to-mesenchymal transdifferentiation is not a source of myofibroblasts or accounts for only a small minority. Other cells that may give rise to interstitial myofibroblasts include resident fibroblasts, endothelial cells, and bone marrow−derived cells. ^,

Several lines of evidence have suggested that this cellular infiltrate contributes to kidney injury and is not merely a consequence of it. In one study, multiple linear regression analysis identified glomerular macrophage infiltration in the remnant kidney as a major determinant of mesangial matrix expansion and adhesion formation between the Bowman capsule and glomerular tufts. Furthermore, depletion of leukocytes by irradiation in rats delayed the onset of glomerular injury after kidney ablative surgery. Several studies have reported amelioration of the cellular infiltrate and kidney injury in the 5/6 nephrectomy model following treatment with the immunosuppressive agent mycophenolate mofetil. One study has found that mycophenolate also lowers P _GC , which may account for some of its renoprotective effects. Several other antiinflammatory interventions have been shown to ameliorate kidney injury after 5/6 nephrectomy. Treatment with the antiinflammatory agent nitroflurbiprofen, also a nitric oxide donor, was associated with moderate kidney protection. Rats treated with a PPAR-γ receptor agonist evidenced significant attenuation of the proteinuria and glomerulosclerosis observed in untreated rats, despite the failure of treatment to lower BP. This kidney protection was observed in association with marked reductions in glomerular cell proliferation, glomerular macrophage infiltration, and renal expression of plasminogen activator inhibitor-1 (PAI-1), as well as TGF-β. The authors speculated that some of these effects may have resulted from the known actions of PPAR-γ receptor activation to antagonize the activities of the transcription factors activator protein-1 (AP-1) and NF-κB.

Administration of a sphingosine-1-phosphate (S1P) receptor agonist, a novel immunosuppressant that inhibits the egress of T cells and B cells from lymph nodes, attenuated the increase in chemokine receptor expression observed in untreated rats and tended to normalize regulated on activation, normal T cell−expressed and secreted (RANTES) and MCP-1 gene expression. Glomerular and interstitial inflammation, as well as fibrosis, were also reduced. Overexpression of the antiinflammatory cytokine interleukin-10 (IL-10) in rats was associated with reduced interstitial inflammation and lower levels of MCP-1, RANTES, interferon-γ (IFN-γ), and IL-2 expression, as well as attenuation of proteinuria, glomerulosclerosis, and TIF. Finally, overexpression of the gene for angiostatin, an antiangiogenic factor that also inhibits leukocyte recruitment, as well as neutrophil and macrophage migration, was associated with the inhibition of macrophage and T cell infiltrates in glomeruli, along with the interstitium, reduced MCP-1 expression, and attenuation of glomerulosclerosis and interstitial fibrosis. Taken together, these findings strongly support the hypothesis that in addition to direct glomerular cell injury, glomerular hemodynamic adaptations to nephron loss provoke a complex series of proinflammatory and profibrotic responses that further contribute to kidney damage. Treatments that antagonize the mediators of these responses may therefore be of benefit in slowing the rate of CKD progression.

Abnormal excretion of protein in the urine is the hallmark of experimental and clinical glomerular disease. Whereas immune complex deposition and resulting inflammation account for abnormal permeability of the glomerular filtration barrier to proteins in glomerulonephritis, studies in rats subjected to extensive kidney ablation have shown loss of glomerular barrier function to proteins of similar molecular size yet in the apparent absence of primary immune-mediated kidney injury or inflammatory response. Sieving studies using dextrans and other macromolecules in rats 7 or 14 days after 5/6 nephrectomy have revealed the loss of both size and charge selectivity of the glomerular filtration barrier. Ultrastructural examination of the remnant kidneys has revealed detachment of glomerular endothelial cells and visceral epithelial cells from the glomerular basement membrane. In addition, protein reabsorption droplets and attenuation of cytoplasm resulting in bleb formation were observed in podocytes. The authors concluded that the altered permselectivity may be caused in part by the separation of endothelial cells from the glomerular basement membrane, allowing access of macromolecules and, in part, to loss of anionic sites in the lamina rara externa, resulting in loss of charge selectivity and detachment of podocytes.

Studies have identified decreased nephrin expression in podocytes as a further mechanism contributing to proteinuria after 5/6 nephrectomy, and in vitro studies have reported a 50% reduction in nephrin expression when podocytes were exposed to cyclic stretching. A direct role for Ang II in modulating glomerular capillary permselectivity has been suggested by the observation of marked increases in urinary protein excretion during infusion of Ang II in normal rats. Although some investigators have attributed this to a direct effect of Ang II on the cellular components of the glomerular filtration barrier, resulting in the opening of interendothelial junctions and epithelial cell disruption, others have shown that the increase in proteinuria may be accounted for almost completely by the associated hemodynamic changes, principally a reduction in Q _A and an increase in filtration fraction. On the other hand, the notion that Ang II may mediate changes in glomerular permselectivity, independent of its effects on glomerular hemodynamics, has been supported by studies in an isolated perfused rat kidney preparation in which infusion of Ang II augmented urinary protein excretion and enhanced the clearance of tracer macromolecules, independently of any change in filtration fraction. Furthermore, Ang II and aldosterone have been shown to reduce nephrin expression in podocytes and may therefore directly affect glomerular permselectivity. ^{, ,}

Proteinuria, long considered simply a marker of glomerular injury, has also been implicated as an effector of injury processes involved in kidney disease progression, especially those resulting in tubulointerstitial fibrosis. A large body of research has demonstrated that proximal tubule epithelial cells cultured in media supplemented with high concentrations of albumin, immunoglobulin G (IgG), or transferrin increase secretion of proinflammatory cytokines. Furthermore, the liberation of these molecules was noted to be predominantly from the basolateral aspect of the cells. This would be in keeping with secretion into the kidney interstitium in vivo, thereby contributing to the development of tubulointerstitial inflammation and fibrosis. Other experiments have found apoptosis in tubule cells exposed to high-molecular-weight plasma proteins but not smaller proteins. Albumin and transferrin exposure also induced complement activation in tubule cells and reduced binding of factor H, a natural inhibitor of the alternative complement pathway. Other investigators have proposed that tubulointerstitial inflammation is provoked by misdirection of protein-rich glomerular filtrate into the interstitium due to the formation of adhesions between the glomerular tuft and Bowman capsule or, in the case of crescentic glomerulonephritis, encroachment of the crescent onto the proximal tubule, resulting in occlusion and tubule degeneration.

Further studies in the 5/6 nephrectomy model have suggested that tubulointerstitial injury may play an important role in the decline of the GFR, especially in the late stages of progressive kidney injury. By examining serial sections of remnant kidneys, the investigators have shown that in association with a doubling in serum creatinine levels, there was a substantial increase in the proportion of glomeruli no longer connected to tubules (atubular glomeruli) or atrophic tubules. Most of these glomeruli were not globally sclerosed, implying that the tubular injury was responsible for the final loss of function in these nephrons. The authors speculated that the absorption of excess filtered protein may play an important role in this tubular injury.

In human studies the magnitude of albuminuria has been identified as an independent risk factor for CKD progression and other adverse outcomes in large meta-analyses, including one with more than 27 million participants. Furthermore, a reduction in albuminuria is strongly associated with a reduction in risk of CKD progression. In a large meta-analysis of 29,979 trial participants, each 30% decrease in mean albuminuria by treatment was associated with an average 27% lower hazard for an endpoint of CKD progression (KRT, GFR<15 mL/min/1.73 m ² or doubling of serum creatinine), providing strong evidence for the use of change in albuminuria as a surrogate endpoint for kidney-protective efficacy in clinical trials.

Taken together, the evidence from experimental and clinical studies provides support for the hypothesis that impaired glomerular permselectivity results in excessive filtration of proteins and/or protein-bound molecules that contribute to kidney damage. In addition, the close association between the severity of proteinuria and kidney prognosis implies that reduction of proteinuria should be regarded as an important independent therapeutic goal in clinical strategies seeking to slow the rate of progression of CKD. The mechanisms and consequences of proteinuria are discussed in detail in Chapter 29 .

Together with secondary FSGS, tubulointerstitial fibrosis constitutes a major component of the progressive kidney injury observed in CKD. Tubulointerstitial fibrosis is characterized by inflammatory cell and fibroblast infiltration, accumulation of extracellular matrix, tubule cell loss, and rarefaction of peritubular capillaries. Fibrogenesis starts at small sites of inflammation and then expands if a profibrotic milieu persists. The inflammatory infiltrate is composed of lymphocytes, macrophages, dendritic cells, and mast cells. Lymphocytes are recruited early in the process, and their importance is highlighted by the protection from fibrosis observed in Rag-2 null mice, which lack B and T lymphocytes. Monocytes are recruited and transdifferentiated into macrophages and fibrocytes. The profibrotic role of macrophages is illustrated by observations that the extent of macrophage accumulation correlates closely with the severity of fibrosis, and macrophage depletion attenuates fibrosis. Nevertheless, it has been proposed that alternatively activated M2 macrophages exert antiinflammatory actions, and the infusion of cells enriched for M2 macrophages has been shown to reduce kidney fibrosis in mice. Myofibroblasts are the chief source of extracellular matrix production, and the accumulation of interstitial myofibroblasts is central to the pathogenesis of tubulointerstitial fibrosis.

There has been considerable controversy regarding the cellular origins of interstitial myofibroblasts in tubulointerstitial fibrosis. Different investigators have identified resident fibroblasts, transdifferentiation from tubule epithelial cells and endothelial cells, ^, bone marrow−derived fibrocytes, and pericytes as possible sources. Fate-mapping studies have subsequently indicated that pericytes are the predominant source of myofibroblasts and that epithelial-to-mesenchymal transdifferentiation (EMT) is not a source of myofibroblasts or accounts for only a minority. Further studies have helped resolve the controversy by showing that TECs undergo partial EMT, expressing both epithelial and mesenchymal markers, but remain attached to the tubular basal membrane. Moreover, prevention of EMT substantially ameliorated kidney fibrosis in animal models, confirming that partial EMT of TECs contributes to the pathogenesis of kidney fibrosis. ^,

The excess extracellular matrix is composed largely of collagen types I and II, as well as fibronectin. Investigators have sought to identify the relative contribution of different cell types to the production of collagen I using cell type−specific knockouts. Experiments have revealed that bone marrow−derived cells (fibrocytes) contribute 38% to 50% of collagen deposition in a unilateral ureteral obstruction model. Furthermore, initial collagen I synthesis was attributable to resident mesenchymal fibroblasts and was beneficial in preserving kidney function in this model, whereas bone marrow−derived cells were responsible for later collagen I production that did not affect kidney function. TECs were found not to contribute directly to collagen I production. Bone marrow−derived cells were similarly shown to contribute substantially to collagen I production in an adenine-induced nephropathy model and, in this case, greater deposition of collagen was associated with lower GFR at all time points. Targeting of bone marrow−derived cell collagen I production therefore represents an attractive therapeutic option to ameliorate kidney fibrosis.

Experimental evidence suggests that fibrosis starts focally with the development of a fibrogenic microenvironment that triggers fibroblast activation, termed a “fibrogenic niche.” Matrix accumulation has been proposed to commence with the appearance of collagen nucleators in the interstitial fluid that acts as a scaffold for the deposition of fibrillar collagens. Fibroblasts use collagen fibrils as a scaffold to move through damaged tissue along chemoattractant gradients. Fibrogenesis is promoted by the expression of key growth factors, principally TGF-β. The role of tissue proteases in tubulointerstitial fibrosis is complex and has not been fully characterized. Whereas matrix metalloprotease (MMP) types 2 and 9 degrade collagen type IV (and possibly type I and III) in vitro, they do not consistently abrogate TIF in vivo. ^, Indeed, in some models, MMPs appear to exert profibrotic effects.

Rarefaction of peritubular capillaries is a hallmark of tubulointerstitial fibrosis. In the early stages, capillaries may be damaged by transient ischemia that promotes apoptosis. Further capillary loss is attributable to an imbalance between proangiogenic and antiangiogenic factors ^, or loss of peritubular endothelial cells through endothelial-mesenchymal transdifferentiation. Detailed studies have identified changes in peritubular capillary endothelium in animal models of CKD, including the development of subendothelial spaces, loss of fenestrations, and generation of caveolae and vesicles, as well as increased permeability and degeneration of the microvascular tree. Tissue hypoxia results from the rarefaction of peritubular capillaries and accumulation of extracellular matrix, requiring oxygen to diffuse over greater distances to reach cells. Hypoxia contributes to tubulointerstitial fibrosis by promoting EMT and apoptosis of tubule cells, as well as fibroblast activation and extracellular matrix production. ^, The importance of hypoxia in provoking interstitial (and glomerular) injury has been demonstrated by studies in which induction of hypoxia-inducible factor, a key mediator of protective responses to hypoxia, was associated with amelioration of proteinuria, glomerulosclerosis, and tubulointerstitial fibrosis, as well as decreased macrophage infiltration and expression of type IV collagen and osteopontin. ^, Tubulointerstitial inflammation and fibrosis are discussed in more detail in Chapter 37 .

TGF-β is associated with chronic fibrotic states throughout the body and is the central mediator of kidney fibrosis. Three isoforms of TGF-β have been described in mammals, but the focus of most research has been on TGF-β1. The active form of TGF-β is a dimer that binds to a transmembrane receptor, which is a heterodimer composed of TGF-β receptor types I and II. Intracellular signaling occurs via Smad-dependent and Smad-independent pathways. Smad2 and Smad3 have been identified as the most important intracellular mediators of the actions of TGF-β, with Smad3 having profibrotic effects and Smad2 proposed to have counterregulatory antifibrotic effects. In response to TGF-β receptor signaling, Smad2 and Smad3 are phosphorylated and form a complex with Smad4 that translocates into the nucleus and binds to DNA to modulate the transcription of multiple target genes. Smad3 has been shown to promote transcription of multiple genes involved in fibrosis including PAI-1, proteoglycans, integrins, CTGF, TIMP-1, and collagen types 1, 5, and 6. SMAD 3 also plays a role in EMT. In addition, TGF-β/Smad3 modulates the production of micro RNA molecules that promote fibrosis; levels of profibrotic miR-21, miR-192, and miR-433 are increased, whereas antifibrotic miR-29 and miR-200 are decreased. Smad4 is a key modulator of the profibrotic actions of TGF-β and reduces the profibrotic actions of Smad3 by inhibiting its binding to the promoter regions of profibrotic genes. Smad7 is an important negative feedback regulator of TGF-β signaling, which acts by preventing the recruitment and phosphorylation of Smad2 and Smad3. Smad7 also increases the expression of inhibitor of kappa B (IκBα), an inhibitor of NF-B, and may therefore be important in mediating the antiinflammatory actions of TGF-β. Finally, Smad signaling may also be activated by TGF-β−independent mechanisms that are relevant to CKD progression including Ang II and advanced glycation end products (AGEs).

In vitro TGF-β elicits overproduction of ECM constituents by mesangial cells, and its expression is increased in several experimental models of kidney disease. These include diabetic nephropathy, anti−Thy-1 glomerulonephritis, adriamycin-induced nephropathy, and chronic allograft nephropathy, as well as human glomerulonephritis, ^, HIV nephropathy, diabetic nephropathy, and chronic allograft nephropathy. The role of TGF-β in kidney fibrosis has been further illustrated by experiments in which transfection of the gene for TGF-β into one renal artery produced ipsilateral kidney fibrosis. In 5/6 nephrectomized rats, a twofold to threefold increase in remnant kidney mRNA levels for TGF-β was observed, and in situ hybridization revealed elevations in TGF-β mRNA throughout the glomeruli, tubules, and interstitium. Treatment with an ACE inhibitor or ARB resulted in substantial kidney protection and prevented upregulation of TGF-β. ^, Furthermore, in rats treated with an ACE inhibitor or ARB, the extent of glomerulosclerosis correlated closely with remnant kidney TGF-β mRNA levels.

Several interventions that inhibit the effects of TGF-β have been shown to afford kidney protection in animal models of kidney disease. Transfection of the gene for decorin, a naturally occurring inhibitor of TGF-β, into skeletal muscle limited the progression of kidney injury in anti−Thy-1 glomerulonephritis. Administration of anti−TGF-β antibodies to salt-loaded, Dahl salt−sensitive rats ameliorated the hypertension, proteinuria, glomerulosclerosis, and interstitial fibrosis typical of this model. Treatment with tranilast (N-[3′,4′-dimethoxycinnamoyl]-anthranilic acid; Pharm Chemical, Shanghai Lansheng, Shanghai, China), an inhibitor of TGF-β−induced ECM production, significantly reduced albuminuria, macrophage infiltration, glomerulosclerosis, and interstitial fibrosis in 5/6 nephrectomized rats. Transfer of an inducible gene for Smad 7, which blocks TGF-β signaling by inhibiting Smad 2/3 activation, inhibited proteinuria, fibrosis, and myofibroblast accumulation after 5/6 nephrectomy. Two weeks of treatment with a polyamide compound designed to suppress transcription of the TGF-β gene significantly reduced proteinuria and prevented upregulation of TGF-β, CTGF, collagen type I α1, and fibronectin mRNA in the kidney cortex. This also suppressed urinary TGF-β excretion and staining for TGF-β by immunofluorescence in salt-loaded, Dhal salt−sensitive rats. Another fibrogenic molecule, CTGF, has also been observed to be overexpressed in kidney biopsies from persons with a variety of kidney diseases. The specific induction of CTGF expression by exogenous TGF-β in mesangial cells ^, and fibroblasts, together with the finding that blocking antibodies to TGF-β inhibits increased CTGF expression in mesangial cells exposed to high glucose concentrations, suggests that CTGF may serve as a downstream mediator of the profibrotic effects of TGF-β. Further experiments have shown that the fibrotic effects of TGF-β in the remnant kidney are mediated, at least in part, by the induction of the micro RNAs miR-21, miR-192, and miR-433 via Smad 3. In vitro overexpression of miR-192 promotes, but inhibition of miR-192, attenuates, TGF-β−induced production of collagen type I in rat tubule cells.

It is clear that interventions to inhibit the multiple mechanisms of kidney fibrosis show promise as treatments for slowing the progression of CKD. However, despite the promising results of many preclinical studies inhibiting the expression or actions of TGF-β discussed previously, a phase 2 randomized trial of anti−TGF-β ₁ monoclonal antibody therapy—added to RAAS inhibitor treatment in persons with diabetic kidney disease, GFR of 20 to 60 mL/min/1.73 m ² , and urine polymerase chain reaction (PCR) value higher than 800 mg/g—found no benefit with respect to change in serum creatinine levels after a median of 315 days of therapy. Secondary analyses also found no differences in proteinuria or biomarkers of kidney injury between groups receiving anti−TGF-β ₁ antibody versus placebo.

This negative result may have been in part attributed to the relatively short trial duration, inadequate dosing of the antibody, or selection of participants with relatively severe and advanced nephropathy. Other possible explanations include redundancy in fibrotic mechanisms and the fact that TGF-β has potentially beneficial antiinflammatory actions. Nevertheless, it is likely that the development of interventions to inhibit kidney fibrosis will remain an area of active research and result in multiple potential novel therapies. ^,

Micro RNAs (miRNAs) are small noncoding RNAs that have been shown to have important posttranscriptional gene regulatory functions. Attention has been focused on the potential role of miRs to modulate kidney fibrosis. miR-21 and miR-214 have been shown to be upregulated in several experimental models of CKD, ^, and miR-21 has been reported to be upregulated in human transplant kidneys with nephropathy. Deletion of miR-21 or treatment with anti-miR-21 oligonucleotides was associated with amelioration of interstitial fibrosis in animal models. Similarly, deletion of miR-214 and treatment with anti−miR-214 were each associated with attenuation of interstitial fibrosis in the UUO model. Importantly, the effects of miR-214 appear to be independent of TGF-β signaling, and TGF-β blockade had an additive antifibrotic effect with miR-214 deletion. miR-214 has further been demonstrated to suppress autophagy, an essential process that cells use to degrade damaged organelles and protein aggregates in lysosomes. In mouse models of type 1 diabetes, impaired autophagy in the kidneys was associated with elevated miR-214 expression, as well as increased hypertrophy and inflammation. Treatment with anti-miR-214 and proximal tubule-specific miR-214 gene knock out both improved autophagy and reduced diabetes-related kidney hypertrophy. Furthermore, miR-214 gene knock out ameliorated albuminuria. As discussed previously, TGF-β/Smad3 signaling increases the production of profibrotic miRs and decreases the production of antifibrotic miRs. Multiple other miRs have been implicated in the pathogenesis of different forms of AKI and CKD. ^,

On the other hand, some miRs may mediate kidney-protective effects. Expression of miR-204-5p was observed to be decreased in the kidneys of persons with hypertensive and diabetic kidney disease. Furthermore, in a mouse model of hypertensive CKD, Mir204 gene knock out exacerbated albuminuria, renal interstitial fibrosis, and arterial disease and in a model of diabetes, treatment with anti-miR-204-5p exacerbated albuminuria and cortical fibrosis.

Studies have identified that miRs can be transferred between cells in extracellular vesicles to exert autocrine or paracrine effects. MicroRNAs may prove to be useful biomarkers to characterize or monitor CKD progression mechanisms and the potential to administer beneficial miRs or antagonize the effects of pathogenic miRs with antagomirs represents a further potential therapeutic strategy for CKD.

Senescence describes a cell phenotype characterized by permanent cell-cycle arrest. It may be induced in many cell types by a range of stressors including DNA damage, mitochondrial damage, epigenetic factors, oncogene activation, oxidative stress, mechanical stress, and factors secreted by other senescent cells. Importantly, senescent cells remain viable and produce a characteristic secretome termed senescence-associated secretory phenotype (SASP), composed of proinflammatory cytokines, chemokines, growth factors, matrix metalloproteinases, and microRNAs that promote inflammation and fibrosis and are proposed to contribute to the pathogenesis of a wide range of chronic diseases. All types of kidney cells may become senescent, but kidney damage and fibrosis are most closely associated with senescent tubule epithelial cells. Senescent tubule cells have been identified in several experimental models of CKD. In the 5/6 nephrectomy and adriamycin nephropathy models, overexpression of C-X-C motif chemokine receptor 4 (CXCR4) in these models was associated with increased cell senescence and interstitial fibrosis, whereas knock-down of CXCR4 ameliorated both. The UUO model is also associated with increased tubule epithelial cell senescence but treatment with the senolytic drug, ABT-263 (also known as navitoclax), which induces apoptosis of senescent cells, resulted in an increase in kidney fibrosis, suggesting that cell senescence may be important for limiting kidney damage during ongoing injury. In contrast, administration of ABT-263 after relief of the ureteric obstruction resulted in decreased fibrosis, implying that cell senescence impaired repair after the injury phase. Similarly, administration of ABT-263 after ischemia-reperfusion injury was associated with reduced tubule cell senescence and less interstitial fibrosis. In humans, senescent cells have been identified in kidneys from persons with ureteric obstruction, glomerulonephritis, CKD, and failing transplants. In persons with primary focal and segmental glomerulosclerosis, the proportion of tubule cells expressing markers of cell senescence was associated with subsequent rate of GFR decline. Multiple senolytic drugs and others that antagonize SASP have shown kidney-protective benefits in experimental models, and several are being tested in clinical trials.

CKD is associated with increased oxidative stress that likely contributes to the progression of kidney damage, as well as the pathogenesis of the associated cardiovascular disease. Superoxide is the primary reactive oxygen species (ROS) that accounts for oxidative stress. Its major source is production by nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, and it is removed (by conversion to hydrogen peroxide by superoxide dismutase (SOD). Following 5/6 nephrectomy, significant upregulation of NADPH oxidase and downregulation of SOD were observed in the liver and kidneys, indicating that the increase in superoxide is due to a combination of increased production and decreased removal. Similarly, in a mouse model of proteinuric CKD induced by the administration of doxorubicin (ADR), NADPH oxidase was upregulated, and extracellular (EC-SOD) was downregulated. Furthermore, EC-SOD knockout mice showed greater albuminuria and more kidney fibrosis after doxorubicin than wild-type mice, confirming that the antioxidant effects of EC-SOD protect against kidney damage. Examination of human kidney biopsies confirmed that EC-SOD is similarly downregulated in human CKD. BP was elevated, and nitrotyrosine levels were increased, whereas urine nitric oxide metabolites were decreased, observations consistent with increased nitric oxide inactivation by superoxide. The effects of increased levels of ROS are further compounded by the reduced abundance and activity of antioxidant enzymes (e.g., catalase, glutathione peroxidase, and glutathione), as well as reduced HDL, apolipoprotein A-1, and thiols.

Adverse consequences of oxidative stress that may contribute to CKD progression include hypertension (due to inactivation of nitric oxide and oxidation of arachidonic acid to generate vasoconstrictive isoprostanes), inflammation (due to activation of NF-κB), fibrosis, and apoptosis, as well as glomerular filtration barrier damage. Inflammation may in turn increase oxidative stress due to the generation of ROS by activated leukocytes, thus establishing a vicious cycle of oxidative stress and inflammation.

Other factors that may contribute to increased ROS production in CKD include Ang II, reduced production of NO, and hypertension. The contribution of cellular injury due to oxidative stress in kidney damage is further illustrated by studies investigating the role of aldehyde dehydrogenase 2 (ALDH2), a mitochondrial enzyme that protects against ROS-induced accumulation of toxic aldehydes. ALDH2 expression was found to be decreased in human kidneys with CKD and in mouse kidneys after damage induced by aristolochic acid (AA) exposure. Furthermore, mutation-induced reduction in ALDH2 expression resulted in increased fibrosis after AA exposure, whereas overexpression of ALDH2 in proximal tubule cells protected them against developing a profibrotic phenotype after exposure to TGF-β. These findings may be particularly relevant in East Asia, where 35% to 45% of populations carry a deactivating mutation in the ALDH2 gene.

The importance of ROS in the progression of CKD has been shown by experiments in which antioxidant therapies including melatonin, niacin, omega-3 fatty acids. and n-acetylcysteine have reduced oxidative stress and ameliorated kidney damage in the 5/6 nephrectomy and other animal models. On the other hand, treatment with tempol, a SOD mimetic, reduced plasma malondialdehyde levels and the number of superoxide-positive cells but did not reduce overall renal oxidative stress, inflammation, or kidney damage. Specific antioxidant therapies have not yet been shown to be effective in slowing human CKD progression, though several effective kidney-protective drugs are postulated to have antioxidant effects (see discussion of kidney-protective drugs, later).

As the GFR declines, the ability of kidneys to excrete hydrogen ions becomes impaired and a new steady state is achieved that allows excretion of acid at the cost of a persistent metabolic acidosis (see earlier, “Acid-Base Regulation”). Acidosis is present in most persons when the GFR falls below 20% to 25% of normal. Chronic metabolic acidosis has multiple adverse consequences including increased protein catabolism, increased bone turnover, induction of inflammatory mediators, insulin resistance, and increased production of corticosteroids and PTH. ^, These observations make it reasonable to consider whether acidosis may also contribute to progressive kidney damage in CKD. Early experiments found no persisting kidney damage after dietary acid loading in rats with normal kidney function and no kidney protection associated with sodium bicarbonate treatment in the 5/6 nephrectomy model, suggesting that acidosis does not initiate or exacerbate kidney damage.

Later experiments, however, have found that an acid-generating diet of casein protein induces tubulointerstitial injury in rats with normal kidney function, whereas a non−acid-inducing diet of soy protein does not. Furthermore, dietary acid supplementation with (NH ₄ ) ₂ SO ₄ in soy protein−fed rats was associated with tubulointerstitial injury. Similarly, in the 5/6 nephrectomy model, a casein-rich diet was associated with metabolic acidosis, a progressive decline in the GFR, and increasing albuminuria, whereas a soy protein diet was not.

Dietary acid supplementation with (NH ₄ ) ₂ SO ₄ in soy protein−fed rats provoked a decline in the GFR and albuminuria. Treatment with sodium bicarbonate or Ca(HCO ₃ ) ₂ , but not sodium chloride, was kidney protective in the casein-fed rats but only if the resultant hypertension—in sodium bicarbonate but not Ca(HCO ₃ ) ₂ −treated rats—was adequately treated. Furthermore, when rats were subject to 2/3 rather than 5/6 nephrectomy, a level of kidney mass reduction not associated with metabolic acidosis, microdialysis identified tissue acid accumulation in muscle and kidney that correlated with the subsequent GFR decline. Amelioration of tissue acid accumulation by a low acid-generating diet or alkali supplementation was associated with abrogation of the GFR decline, whereas dietary acid supplementation was associated with exacerbation of the GFR decline. Treatment with calcium citrate has also been shown to improve acidosis and reduce glomerular, as well as interstitial injury in the 5/6 nephrectomy model. Mechanisms whereby acidosis may contribute to kidney damage after nephron loss include activation of the alternative complement pathway by increased ammoniagenesis, increased levels of plasma and kidney Ang II, and induction of ET, along with aldosterone production.

Observational clinical studies have identified acidosis as an independent risk factor for CKD progression, ^, but to date only relatively small studies have investigated the kidney-protective potential of alkali supplementation in human subjects. In the first randomized study in adults with a creatinine clearance of 15 to 30 mL/min, randomization to treatment of acidosis (serum bicarbonate, 16−20 mmol/L) with sodium bicarbonate was associated with a lower decline in creatinine clearance (1.88 vs. 5.93 mL/min) and lower incidence of end-stage kidney disease (ESKD) (6.5% vs. 33%). The authors conceded, however, that the study was not blinded or placebo controlled. In a second nonrandomized study, treatment of 30 subjects with sodium citrate was associated with reduced urinary excretion of ET-1 and N -acetyl-β- d -glucosaminidase (a marker of tubulointerstitial injury), as well as a lower rate of estimated GFR decline than that observed in 29 untreated controls who were unwilling or unable to take sodium citrate. One study has reported kidney-protective effects following sodium bicarbonate treatment in early CKD. In a randomized, placebo-controlled trial in subjects with a mean estimated GFR of 75 mL/min/1.73 m ² , treatment with sodium bicarbonate for 5 years was associated with a slower reduction in the estimated GFR (derived from plasma cystatin C measurements) than placebo or treatment with sodium chloride.

Further studies have reported that correction of acidosis with a diet rich in fruit and vegetables is as effective in ameliorating kidney damage in early (CKD stage 1 or 2) ⁴⁹² and more advanced (CKD stage 4) disease. In a further trial, people with CKD stage 3 and a serum bicarbonate level of 22 to 24 mmol/L were randomized to oral bicarbonate supplementation, a diet rich in fruit and vegetables, or usual care. All participants received treatment with RAAS inhibition, and SBP was controlled to <130 mm Hg. Both interventions achieved an increase in serum bicarbonate levels and were associated with a decrease in urinary angiotensinogen. After 3 years, both interventions were associated with less albuminuria and GFR decline than the usual care group. A meta-analysis of 6 randomized studies of dietary interventions for improving acidosis that included 644 participants with eGFR 15 to 40 mL/min/1.73 m ² and baseline serum bicarbonate 14 to 24 mmol/L reported an increase in serum bicarbonate concentration (mean difference 2.98, 95% CI 0.77–5.10 mmol/L) difference and eGFR (mean difference 3.16, 95% CI 0.24–6.08 mL/min/1.73 m ² ) versus controls.

On the other hand, a randomized trial of treatment with a hydrochloric acid binder, ververimer, in persons with eGFR 20 to 40 mL/min/1.73 m ² and serum bicarbonate concentration of 12 to 20 mEq/L at baseline found no difference in the combined primary endpoint of KRT, eGFR decline ≥40%, or death due to kidney failure, though the achieved difference in serum bicarbonate concentration was only approximately 1 mEq/L, which may have been inadequate to achieve a difference in outcomes.

Bicarbonate supplementation and/or dietary intervention is recommended to prevent bicarbonate levels lower than 18 mEq/L in persons with CKD, but further studies are required to investigate whether it is beneficial in the setting of less severe acidosis. ^,

As noted, Ang II plays a central role in the glomerular hemodynamic adaptations observed after kidney mass ablation. Ang subtype 1 receptors are, however, distributed on many cell types in the kidney including mesangial, glomerular epithelial, endothelial, tubule epithelial, and vascular smooth muscle, cells suggesting multiple potential actions of Ang II in the kidney. Experimental studies have revealed several nonhemodynamic effects of Ang II that may be important in CKD progression ( Fig. 50.8 ). In isolated perfused kidneys, the infusion of Ang II results in loss of glomerular size permselectivity and proteinuria, an effect that has been attributed to both the hemodynamic effects of Ang II, resulting in elevations in P _GC , and a direct effect of Ang II on glomerular permselectivity. Furthermore, overexpression of angiotensin subtype 1 receptors on podocytes resulted in albuminuria and FSGS in the absence of hypertension in transgenic rats. These effects may be explained in part by observations that Ang II-induced apoptosis and decreased nephrin expression in podocytes.

Schematic depicting the central role of angiotensin II through hemodynamic and nonhemodynamic effects in the pathogenesis of progressive renal injury and fibrosis following nephron loss.

ECM, Extracellular matrix; m ϕ , macrophage; PAI-1, plasminogen activator inhibitor-1; P _GC , glomerular capillary hydraulic pressure; TGF-β, transforming growth factor-β).

From Taal MW, Brenner BM. Renoprotective benefits of RAS inhibition: from ACEI to angiotensin II antagonists. Kidney Int. 2000;57:1803−1817.

In vitro, Ang II has been shown to stimulate mesangial cell proliferation and induce expression of TGF-β, resulting in increased synthesis of ECM. In vivo, transfection of rat kidneys with human genes for renin and angiotensinogen resulted in glomerular ECM expansion within 7 days. Ang II also stimulates the production of PAI-1 by endothelial cells and vascular smooth muscle cells and may therefore further increase the accumulation of ECM through inhibition of ECM breakdown by MMPs that require conversion to an active form by plasmin. Other reports have indicated that Ang II may directly induce the transcription of a variety of cell adhesion molecules and cytokines, as well as activating the transcription factor NF-κB and directly stimulating monocyte activation. Ang II infusion provoked upregulation of COX-2 expression in rats that was not dependent on BP elevation, and 5/6 nephrectomized rats evidenced Ang II-dependent upregulation of interstitial COX-2 expression.

In other experiments, Ang II infusion has been shown to induce interstitial macrophage infiltration and increased expression of MCP-1 and TGF-β, effects that were sustained for up to 6 days after cessation of the infusion. In cell culture and animal experiments, Ang II has been shown to promote tubule cell EMT by suppressing expression of miR-429, which inhibits EMT. Finally, Ang II may have fibrogenic effects via mineralocorticoids (see later). Interestingly, Ang II may also have antifibrotic effects via the angiotensin subtype 2 receptor (AT ₂ ). Ang II appears to upregulate AT ₂ receptor expression via an AT ₂ receptor−dependent mechanism after 5/6 nephrectomy, and treatment with an AT ₂ receptor antagonist exacerbates kidney damage and increased renal PAI-1 expression. Furthermore, overexpression of AT ₂ receptors in transgenic mice was associated with reduced albuminuria, as well as decreased glomerular expression of PDGF-BB chain and TGF-β after 5/6 nephrectomy.

Observations that aldosterone stimulates collagen synthesis in the myocardium and that spironolactone treatment affords survival benefits in addition to that achieved with ACE inhibitor alone in persons with heart failure provided initial impetus to studies investigating the potential role of aldosterone in kidney fibrosis. In the remnant kidney model, adrenal hypertrophy and markedly elevated plasma aldosterone levels have been reported. Furthermore, the administration of exogenous aldosterone during inhibition of the RAAS with combination of ACE inhibitor–ARB therapy in the 5/6 nephrectomy model negates the kidney-protective effects of the latter. Further evidence of the role of aldosterone has been provided by experiments in which rats subjected to adrenalectomy after 5/6 nephrectomy received replacement glucocorticoid but not mineralocorticoid therapy, resulting in less severe kidney injury than rats with intact adrenal glands. In persons with CKD, higher plasma aldosterone concentration was associated with lower eGFR and more severe proteinuria. Furthermore, each doubling of baseline serum aldosterone concentration was associated with an 11% increased risk (HR 1.11, 95% CI, 1.04–1.18) of reaching the composite primary endpoint of ESKD or a ≥50% decline in eGFR.

Further exploration of the physiology of aldosterone has revealed that Ang II may not be the main stimulus for aldosterone secretion and that the mineralocorticoid receptor may undergo nonligand activation by regulatory protein Rac family small guanosine triphosphatase (Rac-1), high glucose, and high sodium concentration.

Mechanisms whereby aldosterone and/or mineralocorticoid receptor activation may contribute to kidney damage include hemodynamic effects (see previously), mesangial cell proliferation, apoptosis, hypertrophy and transdifferentiation, podocyte injury, and apoptosis associated with reduced expression of nephrin and podocin, resulting in proteinuria, ^{, ,} proximal tubule cell transdifferentiation and increased production of collagen types III and IV, and increased kidney production of ROS, ^, PAI-1, ^528,529 TGF-β, and CTGF. ^,

Early experimental use of aldosterone receptor blockers in 5/6 nephrectomized rats yielded only modest renoprotective effects, ^, but other studies have found significant amelioration of glomerulosclerosis in 5/6 nephrectomized rats treated with spironolactone, alone or in combination with triple antihypertensive therapy or an ARB. ^{, ,} In some rats, spironolactone was associated with the apparent regression of glomerulosclerosis. Furthermore, the observed kidney protection was associated with inhibition of PAI-1 mRNA expression in the kidney cortex. In another experiment, kidney protection, achieved with combination ACE inhibitor and spironolactone treatment, was associated with abrogation of the increase in mesangial cells and decrease in podocytes observed in untreated rats. Combination treatment also attenuated the increase in expression of type IV collagen, TGF-β, and desmin. Spironolactone has also been shown to ameliorate kidney damage in other experimental models including diabetic nephropathy, ^, Ren2 transgenic rats, radiation nephritis, and stroke-prone hypertension.

The consistent observation of kidney, and in particular glomerular, hypertrophy after kidney mass reduction has prompted the hypothesis that processes involved in, or resulting from, hypertrophy may contribute to progressive kidney injury in CKD. The well-documented observation that kidney and glomerular hypertrophy precede the development of diabetic nephropathy and the finding of a positive association between glomerular size and early sclerosis in rats subjected to kidney mass ablation suggests that hypertrophy may play a direct role in the pathogenesis of glomerulosclerosis. Several clinical observations have also supported an association between glomerular hypertrophy and kidney injury. Oligomeganephronia, a rare congenital condition with a nephron number 25% of normal or less, is characterized by marked hypertrophy of the remaining glomeruli and the development of proteinuria and kidney failure in adolescence, with FSGS as the typical kidney biopsy finding. In children with minimal change disease, a glomerulopathy generally associated with spontaneous remission and lack of progression to kidney failure, investigators have noted an association between glomerular size and the risk of developing FSGS and kidney failure.

Several interventions have been used in experiments to interrupt the development of glomerular hypertrophy after kidney mass reduction and thereby assess its role in kidney disease progression, but they have produced contradictory results. Rats subjected to 5/6 nephrectomy were compared with rats in which 2/3 of the left kidney was infarcted and the right ureter drained into the peritoneal cavity, an intervention that apparently results in decreased renal clearance without compensatory kidney hypertrophy. Micropuncture studies confirmed similar degrees of elevation of P _GC and SNGFR in both models. At 4 weeks, however, the maximal planar area of glomeruli was significantly less, and glomerular injury, as assessed by sclerosis index, was significantly reduced in ureteroperitoneostomized rats versus 5/6 nephrectomized controls. Accordingly, the authors concluded that glomerular hypertrophy is more important than glomerular capillary hypertension in the progression of glomerular injury in this model.

Dietary sodium restriction has also been used to inhibit kidney hypertrophy after 5/6 nephrectomy. Although sodium restriction had no effect on glomerular hemodynamics, glomerular volume was significantly reduced in 5/6 nephrectomized rats fed low versus normal sodium diets. Moreover, urinary protein excretion was lower, and glomerulosclerosis was less severe, in rats on restricted sodium intake. These findings were extended by another study in which the effect of sodium restriction in preventing glomerular hypertrophy and ameliorating glomerular injury was confirmed, but which also found that these benefits were overcome by the administration of an androgen that stimulated glomerular hypertrophy, despite sodium restriction. Glomerular hemodynamics were similar among the groups. On the other hand, treatment with seliciclib, a CDK inhibitor, reduced kidney hypertrophy by 45% after 5/6 nephrectomy but had no effect on kidney damage.

Glomerular hypertrophy may contribute to glomerulosclerosis through a number of different mechanisms. According to the law of Laplace, the increase in glomerular volume could result in an increase in capillary wall tension only if the capillary wall diameter were also increased ( Fig. 50.9 ). Cyclic stretch would then exert stress capable of damaging epithelial, mesangial, and endothelial cells, as described previously. Alternatively, glomerulosclerosis may be viewed as a maladaptive growth response following loss of kidney mass and resulting in excessive mesangial proliferation and extracellular matrix production. The identification of TGF-β as a key mediator of kidney hypertrophy, as well as an important promoter of kidney fibrosis provides an obvious link between kidney hypertrophy and fibrosis.

Illustration of the synergistic effects of changes in transcapillary hydrostatic pressure difference (ΔP, mm Hg) and mean glomerular capillary radius (μm) on the calculated capillary wall tension (dynes/cm).

From Bidani AK, Mitchell KD, Schwartz MM, et al. Absence of progressive glomerular injury in a normotensive rat remnant kidney model. Kidney Int. 1990;38:28–38.

Detailed studies of podocyte hypertrophy have found that impaired podocyte hypertrophy in response to glomerular enlargement is a further mechanism that may contribute to the development of proteinuria and FSGS. Using a transgenic rat model (dominant negative AA-4E-BP1 transgene) that resulted in impaired podocyte hypertrophy, the investigators observed a remarkable linear relationship among gain in body weight, proteinuria, and glomerulosclerosis that was accelerated after uninephrectomy. Dietary caloric restriction that prevented weight gain and glomerular enlargement also prevented proteinuria. Analysis of the kidneys from rats with proteinuria demonstrated a mismatch between glomerular tuft volume and total podocyte volume, and electron microscopy revealed a pulling apart of podocyte foot process, resulting in exposed areas of glomerular basement membrane and adhesions to the Bowman capsule. These findings were extended in the same model by showing that the prevention of glomerular hypertrophy after uninephrectomy by dietary caloric restriction, mTORC1 kinase pathway inhibition (with rapamycin), or treatment with an ACE inhibitor each prevented proteinuria and FSGS. Proliferation of glomerular cells other than podocytes was identified and has been proposed to contribute to the development of FSGS. Moreover, transcriptomic analysis of gene expression in isolated glomeruli from rats that developed FSGS evidenced a similar pattern of gene expression to that identified in glomeruli from human biopsies of persons with progressive CKD, suggesting that similar factors play a role in human CKD progression.

Experimental evidence showing a central role for Ang II in mechanisms of CKD progression through hemodynamic and nonhemodynamic effects has been borne out in randomized clinical trials of ACE inhibitor and ARB treatment in persons with all forms of CKD. ACE inhibitor treatment has been shown to be kidney protective in individual trials of persons with type 1 diabetes and overt nephropathy, microalbuminuria and type 2 diabetes mellitus, and nondiabetic CKD. ^{, , ,} Further trials have found that ARB treatment affords kidney protection in persons with type 2 diabetes and microalbuminuria or overt nephropathy. ^, These findings have been confirmed in several meta-analyses. However, more complete inhibition of the RAAS with combination ACE inhibitor and ARB ^, or combination ARB and direct renin inhibitor treatment was associated with an excess of adverse effects including acute kidney injury (AKI), hypotension, and hyperkalemia, implying that near-complete blockade of the RAAS may be undesirable in many patient groups. In summary, the large body of clinical evidence showing kidney-protective effects of RAAS inhibitor treatment confirms that Ang II is a critical mediator of mechanisms of CKD progression in humans and supports the consensus that RAAS inhibition should be central to treatment strategies for slowing CKD progression. The role of RAAS inhibitor treatment in achieving optimal kidney protection is discussed further in Chapter 54 .

Evidence of the kidney-protective effects of a class of drug developed for the treatment of diabetes, the sodium-glucose cotransporter 2 (SGLT2) inhibitors, has provided further support for importance of glomerular hemodynamic factors in CKD progression. These drugs act by inhibiting the reabsorption of filtered glucose and sodium in the proximal tubule, resulting in glycosuria that reduces hyperglycemia, and modest natriuresis. Importantly, it has been proposed that increased delivery of sodium chloride to the macula densa results in increased tubuloglomerular feedback, which reduces glomerular hyperfiltration by provoking the constriction of afferent arterioles. In rats with streptozotocin-induced diabetes, treatment with an SGLT2 inhibitor resulted in a marked reduction in SNGFR that was inversely proportional to the distal tubular chloride concentration, indicating activation of tubuloglomerular feedback. Similarly, treatment with the SGLT2 inhibitor after 5/6 nephrectomy afforded kidney protection similar to treatment with an ARB and provoked an increase in urinary adenosine excretion, a marker of tubuloglomerular feedback activity, that was inversely related to the severity of tubulointerstitial fibrosis. In persons with type 1 diabetes, treatment with an SGLT2 inhibitor reduced glomerular hyperfiltration to normal levels in conjunction with a decrease in effective renal plasma flow and an increase in renal vascular resistance, observations consistent with preglomerular (afferent arteriole) vasoconstriction. Thus treatment with SGLT2 inhibitors ameliorates glomerular hyperfiltration and (by implication) glomerular hypertension. Landmark clinical trials reported substantial kidney protection when SGLT2 inhibitors were added to RAAS inhibitor treatment in persons with CKD associated with type 2 diabetes or without diabetes and with or without albuminuria. In a meta-analysis of large clinical trials in persons with and without type 2 diabetes, treatment with SGLT2 inhibitors was associated with a significant reduction in risk of the composite primary endpoint of CKD progression (50% or greater decrease in eGFR, a sustained low eGFR, ESKD, or death from kidney failure) (relative risk 0.63; 95% CI, 0.58–0.69), confirming the importance of glomerular hyperfiltration (and glomerular hypertension) in human CKD progression. Nevertheless, several other potential kidney-protective effects of SGLT2 inhibitors have been identified including lowering of systemic BP, reduction of albuminuria, weight loss, amelioration of arterial stiffness, reduction of renal oxygen demand, and induction of hypoxia-inducible factor 1 (HIF1). The kidney-protective effects of SGLT2 inhibitors are reviewed in more detail in Chapter 41 , Chapter 54 .

Several small clinical trials reported an additional reduction of proteinuria by 15% to 54%, BP by approximately 40%, and GFR by approximately 25% when aldosterone receptor blockers were added to ACE inhibitor or ARB treatment. More novel nonsteroidal aldosterone antagonists are also being evaluated. In one randomized trial that enrolled 336 subjects with hypertension, urine ACR of 30 to 599 mg/g, and an eGFR of 50 mL/min/1.73 m ² or higher, treatment with eplerenone for 52 weeks, added to ACE inhibitor or ARB therapy, resulted in a 17.3% decrease in urine ACR, whereas a 10.3% increase was observed in those who received placebo ( P = 0.02). The most compelling evidence of kidney protection has come from trials of a nonsteroidal MRA, finerenone. In the Finerenone in Reducing Kidney Failure and Disease Progression in Diabetic Kidney Disease (FIDELIO-DKD) trial, 5734 persons with type 2 diabetes, GFR 25 to 75 mL/min/1.73 m ² and urine albumin-to-creatinine ratio (ACR) 30 to 5000 mg/g were randomized to treatment with finerenone or placebo, added to RAAS inhibition. Finerenone treatment resulted in a significant reduction in the primary composite endpoint of kidney failure (GFR <15 mL/min/1.73 m ² or KRT), 40% GFR decline, or renal death (HR 0.82; 95% CI, 0.73–0.93). Similarly, The Finerenone in Reducing Cardiovascular Mortality and Morbidity in Diabetic Kidney Disease (FIGARO-DKD) trial enrolled 7437 persons with type 2 diabetes and albuminuria but with a wider range of GFR (25–90 mL/min/1.73 m ² ). In this trial the primary composite endpoint was a major cardiovascular event and the secondary composite outcome was CKD progression (same as the primary endpoint in FIDELIO-DKD). Finerenone treatment significantly reduced the risk of the primary cardiovascular endpoint (HR 0.87; 95% CI, 0.76–0.98) and tended to reduce the risk of the secondary endpoint (HR 0.87; 95% CI, 0.76–1.01). Data from FIDELIO-DKD and FIGARO-DKD were combined in a prespecified pooled analysis showing a significant 23% reduction in the primary composite endpoint of kidney failure, 57% reduction in GFR, or death due to renal causes (HR 0.77, 95% CI 0.67–0.88).

Together, these findings provide strong support for an important role of aldosterone and mineralocorticoid receptor activation, separate from Ang II, in CKD progression mechanisms. Single-cell transcriptomic studies in animal models have provided new insights into the molecular mechanisms of MRA.

Glucagon-like peptide-1 (GLP-1) is a hormone secreted by gastrointestinal tract cells in response to a meal that stimulates glucose-dependent insulin secretion. GLP-1 also slows gastric emptying and acts on the hypothalamus to reduce appetite. GLP-1 receptor agonists (GLP-1RAs) were developed as drugs for the treatment of type 2 diabetes. Subsequent trials showed efficacy in improving glycemic control but also observed substantial weight loss. Following evidence of kidney protection with GLP-1RA treatment from secondary analyses of earlier trials, the “Evaluate Renal Function with Semaglutide Once Weekly” (FLOW) trial was the first to investigate kidney endpoints in persons with CKD. Persons ( n = 3533) with CKD and type 2 diabetes were randomized to treatment with semaglutide or placebo in addition to RAAS inhibitor treatment. The primary endpoint was a composite of kidney failure (KF), 50% decline in estimated glomerular filtration rate (eGFR), or death due to kidney disease or CVD. After a median follow-up period of 3.4 years, the primary composite endpoint of KF, 50% decline in estimated glomerular filtration rate (eGFR), or death due to kidney disease or CVD was reduced by 24% in the semaglutide versus placebo group (HR 0.76; 95% CI, 0.66–0.88). Secondary analysis of data from a trial that observed a 20% reduction in cardiovascular events with semaglutide treatment in persons with obesity but not diabetes reported 22% reduction in the composite kidney endpoint of eGFR <15 mL/min/1.73 m ² , initiation of KRT, death due to kidney disease, eGFR decline of ≥50%, or progression to macroalbuminuria (HR 0.78; 95% CI, 0.63–0.96). In a phase 2 trial of semaglutide treatment in persons with BMI ≥27 kg/m ² , semaglutide treatment reduced albuminuria by 52.1% (95% CI,–65.5 to–33.4; P < 0.0001) compared with placebo.

Kidney protection with GLP-1RA treatment has been attributed to a reduction in risk factors for CKD progression due to weight loss and BP lowering, as well as improvements in dyslipidemia and glycemic control. In addition, GLP-1 receptors are expressed in the kidneys and exert antiinflammatory effects by inhibiting signaling by receptors for advanced glycation end products (RAGE).

Currently recommended kidney-protective drugs therefore impact multiple different common pathway mechanisms of CKD progression. The benefits of SGLT2 inhibitors, MRAs, and GLP-1RAs have been demonstrated to be additive to those of RAAS inhibitors. Further trials are required to investigate whether MRAs and GLP-1RAs afford additional kidney protection when added to combination RAAS inhibitor and SGLT2 inhibitor treatment, but their differing mechanisms of action suggest that this will be the case. For a figure summarizing the impact of kidney-protective drugs on different mechanisms of CKD progression, please see fig. 2 in Chapter 54 .