The Role of the Vascular Niche in Regulating Kidney Tubulo-Interstitial Fibrosis

Abstract

Chronic kidney disease (CKD) is a substantial global health problem with devastating impacts on patients’ morbidity and mortality. Kidney fibrosis, specifically tubulointerstitial fibrosis, is considered the final common pathway in the progression of virtually all forms of CKD. Peritubular capillary rarefaction, which refers to a decrease in peritubular capillary density leading to hypoxic and ischemic conditions, has long been recognized as a hallmark pathologic feature of tubulointerstitial fibrosis and a pivotal biological alteration leading to CKD progression. Conversely, recent literature has challenged this paradigm by proposing that tubulointerstitial fibrosis and CKD progression are closely associated with the upregulation of pro-angiogenic pathways. As such, peritubular capillary rarefaction may be a consequence rather than a cause of tubulointerstitial fibrosis. Furthermore, a growing body of evidence suggests that the microenvironment of the kidney vasculature, which may be referred to as the ‘vascular niche’, is a dynamic entity that regulates vascular homeostasis, key molecular signaling pathways and inflammation. In this review, we detail how the vascular niche may modify the course of various kidney diseases by influencing cell differentiation and the immune response. Understanding the complex interplay between the cellular and molecular components of the vascular niche may eventually lead to the identification of novel therapeutic targets to limit tubulointerstitial fibrosis and halt CKD progression. This could potentially involve modulating the secretion of angiocrine factors, regulating immune cell activity within the vascular niche, or interfering with the transformation of endothelial cells and pericytes into myofibroblasts, which are key players in kidney fibrogenesis.

Introduction

Kidney disease, ranging from acute kidney injury (AKI), chronic kidney disease (CKD) to end-stage-kidney disease (ESKD), represents a global health challenge that impacts more than 850 million people worldwide.^[1] CKD has been recognized as a significant medical and socioeconomic burden. This devastating condition is associated with an increased risk for cardiovascular disease and all-cause mortality.^[2,3] In 2019, Medicare expenditure for combined CKD and ESKD in the United States exceeded US$124 billion.^[4] Unfortunately, ageing populations and the increasing burden of comorbidities are likely to translate to an increased prevalence of CKD in the coming decades. In fact, CKD is projected to be the fifth highest cause of years of life lost globally by 2040.^[5] The annual direct costs of kidney disease are projected to increase by 9.3% between 2022 and 2027, from $372.0 billion to $406.7 billion across 31 countries worldwide.^[6]

Kidney fibrosis, particularly tubulo-interstitial fibrosis (TIF), is the inevitable common endpoint of virtually all progressive kidney diseases, irrespective of the underlying etiology.^[7] It is well-accepted that the development of glomerular and interstitial inflammation precedes TIF,^[8] and is induced originally, as a potentially protective response to various forms of kidney injury.^[9] However, unresolved prolonged inflammation and leukocyte infiltration instigate resident kidney cells to generate excessive profibrotic cytokines and growth factors.^[10] This in turn activates fibroblasts and stimulates maladaptive deposition of extracellular matrix (ECM), leading to TIF and kidney disease progression.^[11] The underlying complex mechanisms that promote kidney inflammation and subsequent fibrosis remain incompletely understood.^[12] Furthermore, it has been recognized that TIF originates in specific foci within the kidney parenchyma.^[13] These foci are typically non-homogeneously distributed and are characterized by distinctive spatially confined profibrotic microenvironments. Understanding the components of these fibrogenic microenvironments and the mechanisms responsible for driving chronic inflammation and fibrosis within them and beyond, is essential for facilitating the development of therapeutics to mitigate the progression of kidney disease.^[14]

The term ‘vascular niche’ refers to the local tissue microenvironment surrounding blood vessels, comprised of capillary endothelial cells, pericytes, fibroblasts, immune cells and tubular epithelial cells. The components of the vascular niche play a crucial role in immune cell recruitment, inflammation and fibrosis. Although a vascular niche may eventually transform into a ‘fibrogenic niche’ (i.e., a profibrotic microenvironment described above), the independent concept of a vascular niche is important because it underscores the fundamental role of the vasculature as a conduit for immune cell trafficking and molecular signaling across the kidney parenchyma. In essence, our understanding of kidney inflammation, repair and fibrosis may be conceptually organized around the biology and pathophysiology of the kidney microvasculature.^[15]

It has been proposed that vascular rarefaction, which refers to a decrease in capillary density leading to hypoxic and ischemic conditions, is a key factor in the development of TIF and CKD progression.^[16] However, the precise mechanisms underlying vascular rarefaction are not fully understood.^[17] For example, global glomerulosclerosis may disrupt the post-glomerular circulation, resulting in peritubular capillary (PTC) rarefaction.^[18] The demise of downstream PTCs in this scenario may be attributable to hypoxia triggering endothelial cell apoptosis and pericyte detachment.^[19,20] Further mechanistic insights are required to fully elucidate these processes. Moreover, there is no current evidence suggesting that new capillaries sprouting from non-sclerotic glomeruli can effectively revascularize the areas previously perfused by sclerosed glomeruli. The organized arrangement of the kidney microvasculature and the relatively poor regenerative ability of PTCs make such an adaptive process unlikely to compensate for extensive glomerulosclerosis.^[21] Furthermore, there is speculation that unique vascular niche microenvironments may also influence the fate of the microvasculature after kidney injury, with previous studies demonstrating, for instance, that the downregulation of angiogenic mediators such as vascular endothelial growth factor (VEGF) induces endothelial cell apoptosis, vascular rarefaction and kidney disease progression.^[22–24]

Conversely, antibody,^[25,26] and aptamer-based^[27] plasma proteomic analyses in CKD patients have identified various novel biomarkers that indicate activated angiogenesis, characterized by the formation of new blood vessels from pre-existing vasculature in response to hypoxia and endothelial injury, supporting the notion of accelerated angiogenesis in the pathogenesis of TIF. In addition, mouse models of kidney disease,^[28–30] described in more detail in subsequent sections, have provided further insights into how increased angiogenesis can contribute to chronic inflammation and kidney fibrosis by promoting macrophage infiltration, endothelial cell changes and immune cell trafficking. Therefore, these data^[25–30] support a contrasting hypothesis, whereby vascular rarefaction is not a cause of kidney TIF, but rather a consequence of it (Figure 1).

Figure 1.

Microvascular rarefaction: a cause or a consequence of kidney tubulointerstitial fibrosis (TIF)?

The purpose of this review is to: 1) introduce the concept of the vascular niche in the kidney parenchyma and its role in tissue homeostasis and remodeling through ‘angiocrine’ mediators; 2) summarize our current understanding of angiogenic and angiostatic activity in the vascular niche; 3) examine the role of the vascular niche in promoting the development and progression of TIF; 4) explore the vascular niche as a potential therapeutic target for TIF to halt the progression of kidney disease.

The Concept of the Vascular Niche in the Kidney Parenchyma

Ecologists have developed the concept of a ‘niche’ to describe the specific characteristics of the natural environment that an organism survives in and the influences that the organism has on its environment.^[31] Similarly, we can refer to the physical and biochemical microenvironments surrounding blood vessels in the kidney parenchyma as the ‘vascular niche.’^[32] A variety of cell populations, including structural cells such as endothelial cells, pericytes and tubular epithelial cells, as well as immune cells such as macrophages, reside within the vascular niche^[33] (Figure 2).

Figure 2.

Components of the vascular niche in the kidney parenchyma.

Endothelial cells and pericytes are the principal structural cells of the vascular niche.^[34,35] Endothelial cells lining capillaries are far from passive perfusion conduits^[36]; they also contribute to the maintenance of resident stem cell homeostasis and guide tissue regeneration by deploying various stimulatory and inhibitory growth factors, chemokines, cytokines and ECM components, collectively referred to as ‘angiocrine’ factors.^[37] Recent advancements in single-cell transcriptomics have uncovered a remarkable degree of endothelial cell diversity within the kidney.^[38,39] This heterogeneity likely reflects the different microenvironmental factors within the glomerular and tubulointerstitial compartments of the kidney, including blood flow, oxygen tension, angiocrine milieu and cellular interactions.^[40] This review primarily focuses on the distinct role of PTC endothelial cells within the vascular niche in kidney health and disease. For instance, unlike glomerular endothelial cells, PTC endothelial cells feature diaphragms that span across their fenestrations, which facilitates their crucial role in regulating vascular permeability.^[41] The PTC endothelium highly expresses the PLVAP (Plasmalemma Vesicle Associated Protein) gene, and its encoded PV-1 protein, which outlines these fenestral diaphragms, distinguishing it from the glomerular endothelium.^[42] Similarly, PTC endothelial cells express higher levels of SLC2A1 (encoding the GLUT1 glucose transporter) compared to other kidney endothelial populations, underscoring the importance of PTCs in maintaining glucose homeostasis.^[43] The PTC endothelium also exhibits elevated expressions of insulin-like growth factor binding protein-3 (IGFBP3) and natriuretic peptide receptor-3 (NPR-3) under basal conditions, whilst von Willebrand factor (vWF) is more markedly expressed in the glomerular endothelium.^[42,43] This distinct transcriptional activity in PTC endothelial cells^[44] needs to be investigated further to elucidate the role of this heterogeneity in kidney health and disease.

Pericytes, on the other hand, are interstitial mesenchymal cells located within the basement membrane in close proximity to endothelial cells.^[45] They were initially recognized for providing structural stability to the microvasculature, regulating cortical and medullary blood flow and participating in erythropoietin production.^[46] However, they are now known to also engage in paracrine crosstalk with neighboring endothelial cells, tubular epithelial cells and immune cells within the vascular niche.^[47] Consequently, they play a central role in maintaining proper endothelial function, microvascular development and maturation, as well as remodeling through pericyte-myofibroblast transition following kidney injury.^[48] Both endothelial cells and pericytes can promote the recruitment of immune cells, such as macrophages, into the vascular niche by secreting various chemokines and cytokines, including monocyte chemoattractant protein-1 (MCP-1) and vascular cell adhesion molecule-1 (VCAM-1).^[49] In turn, pro-inflammatory macrophages can secrete VEGF-A to promote endothelial proliferation and stimulate angiogenesis,^[50] whereas profibrotic macrophages can contribute to tissue remodeling and fibrosis via the macrophage to myofibroblast transition (MMT) pathway.^[51]

The kidney cortex and medulla exhibit distinct cellular and molecular compositions in their vascular niches, largely due to their specialized functions and differing microenvironments.^[44] The medulla is characterized by a state of relative hypoxia and hyperosmolarity compared to the cortex.^[52] Exposure to hypoxic conditions triggers the activation of hypoxia-inducible transcription factor (HIF) dependent pathways, particularly HIF1α expressed by medullary endothelial cells, which upregulates anaerobic glycolysis, as well as HIF2α, involved in the metabolic adaptation to chronic hypoxia.^[53] Similarly, medullary epithelial cells demonstrate a significantly higher capacity for anaerobic glycolytic flux compared to proximal tubular epithelial cells.^[54] The medullary endothelium has also adapted to extreme levels of hyperosmolarity (up to 1400 mOsm/kg in humans) by activating protective mechanisms such as the expression of heat-shock proteins that mitigate protein denaturation induced by elevated urea concentration.^[55] A single-cell transcriptome study in C57BL/6J mice exposed to water deprivation demonstrated that medullary endothelial cells also participate in the import and synthesis of inert organic osmolytes, such as glucose-derived polyols, to safeguard against hyperosmolarity-induced cell damage.^[42] Furthermore, medullary pericytes possess a higher proportion of circumferential processes wrapping around vessels rather than the longitudinal processes found in cortical pericytes.^[56] This difference in pericyte morphology and the higher density of medullary pericytes may suggest a greater capacity for regulating microvascular blood flow in this region. This is crucial for maintaining oxygen and nutrient supply in the face of physiological challenges like ischemia and might lead to distinct patterns of capillary constriction or dilation following kidney injury.^[56,57]

Lastly, kidney lymphatic networks, with their specialized endothelium and ability to form new lymphatic vessels from pre-existing ones in response to injury i.e. lymphangiogenesis, have recently been proposed to function as a specialized ‘vascular’ niche within the kidney.^[58] Under basal conditions, lymphatics assist PTCs in preventing the accumulation of interstitial fluid and solutes that may have escaped reabsorption by PTCs.^[59] Although the immune regulatory function of kidney lymphangiogenesis in disease states has been increasingly recognized,^[60,61] there are pronounced gaps in our understanding of the specific roles and mechanisms of the lymphatic vascular niche in kidney disease.^[62,63]

In summary, residents of the vascular niche maintain the steady state of the microenvironment under normal physiological conditions by establishing a delicate balance between the formation of new capillaries and the pruning of existing ones, ensuring optimal perfusion and meeting the metabolic demands of the tissue.^[64] This dynamic remodeling of the microvasculature through capillary turnover regulates tissue homeostasis and is accomplished through the deployment of angiocrine factors.^[65] However, dysfunctional resident cells can also reshape the vascular niche in response to tissue injury, for instance, by modifying tight junction proteins within endothelial cells,^[66] by releasing chemokines and cellular adhesion molecules,^[67] by creating an imbalance between pro-angiogenic and anti-angiogenic factors leading to dysregulated angiogenesis,^[68] and by promoting pericyte/macrophage to mesenchymal transition in fibrosis.^[69,70] Therefore, the vascular niche actively participates in the regulation of kidney inflammation, repair and fibrosis and consequently, plays a pivotal role in the development and progression of kidney disease.^[71]

The Intricate Balance of Angiogenic and Angiostatic Factors in the Vascular Niche

The integrity of the peritubular microvascular network and crosstalk between the various components of the vascular niche are dependent on the balance between the local expression of pro-angiogenic and anti-angiogenic factors, and a disruption of this balance has been identified in various kidney diseases^[37,44] (Table 1). The glomerular and peritubular vascular niches are not isolated entities but rather, are intricately connected compartments.^[92] As such, although the peritubular vascular niche is the review’s central focus, its complex interplay with the angiocrine activities and cellular processes of the glomerular vascular niche in the regulation of kidney TIF will be briefly described in the remainder of this review.

Table 1 |.

Proangiogenic and anti-angiogenic factors within the vascular niche: role in kidney microvasculature and TIF/kidney disease

Angiocrine factor	Receptor/signaling pathway	Sites of kidney expression	Role in kidney microvasculature	Role in TIF/kidney disease	Citations
VEGF-A	VEGFR2	Tubular epithelial cells Podocytes	Promotes endothelial cell migration and proliferation; maintains normal PTC network	Increased tubular VEGF-A can lead to fibrosis by influencing cell-matrix adhesion and fibroblast migration	24, 72
sVEGFRl	VEGF-A	Endothelial cells Podocytes Macrophages	Prevents activation of VEGFR2 by binding and sequestering VEGF-A; reduces angiogenesis	Potential protective role against fibrosis; reduces macrophage infiltration in response to increased VEGF in DKD	22, 73
EphrinB2	EphB4	Endothelial cells Pericytes	Modulates angiogenesis by controlling the internalization and localization of VEGFR2	EphrinB2 reverse signaling and mitigates pericyte-myofibroblast transition and myofibroblast activation, limiting fibrosis	74, 75
Semaphorin 3A	NRP1	Tubular epithelial cells Podocytes	Negative regulatory role in angiogenesis; inhibits VEGF-A-induced VEGFR2 signaling	Contributes to fibrosis by promoting epithelial-mesenchymal transition and JNK pathway and fibroblast activation	76
PEDF	PEDFR	Endothelial cells Podocytes	Cleaves VEGFR2 by enhancing γ-secretase at the transmembrane region; inhibits angiogenesis	Suppresses fibrogenesis by inhibiting TGF-β activity, WNT/β-catenin pathway, and fibroblast activation	77, 78
ANG-1	TIE2	Tubular epithelial cells Podocytes	Stabilizes vascular endothelium, and maintains vascular integrity and barrier function	Promotes antiinflammatory response by downregulating ICAM1/VCAM1; decreased ANG-1 levels lead to fibrosis	68, 79, 80
ANG-2	TIE2	Weibel Palade bodies of endothelial cells	Counterregulatory to ANG-1, destabilizing the endothelium and promoting angiogenesis	ANG-2/ANG-1 ratio elevated in CKD and fibrosis; ANG-2 promotes endothelial apoptosis and macrophage infiltration	28, 81
VEPTP	ANG/TIE2	Endothelial cells	Negative regulator of the ANG/TIE2 signaling pathway	VEPTP inhibition activates eNOS, excludes FOXO1 transcription factor, reduces profibrotic gene expression	82, 83
TGF-β	ALK1: angiogenic ALK5: angiostatic	Tubular epithelial cells Podocytes	Various aspects of endothelial cell function, including proliferation and apoptosis	Crucial role in fibrosis; stimulates ECM production, fibroblast activation, and Endo-MT and inhibits ECM degradation	84, 85
LRG1	Endoglin TGF-βR-ll	Endothelial cells Tubular epithelial cells	Promotes TGF-β-mediated proangiogenic activity via ALK1-Smad-1/5/8 signaling pathway	Tubular overexpression of LRG1 promotes TIF by potentiating TGF-β/Smad3 signaling in tubular cells	86, 87
VASH1	Unidentified	Endothelial cells Podocytes	Angiostatic; inhibits sprouting angiogenesis and promotes vascular maturation/quiescence	Protective factor against fibrosis by mitigating oxidative stress and inhibiting fibroblast activation	88, 89
VASH2	Unidentified	Tubular epithelial cells in response to injury	Proangiogenic; low levels in normal kidneys and upregulated in ischemia-reperfusion injury	Potentially contributes to fibrosis by promoting epithelial- mesenchymal transition and TGF-β signaling	90, 91

ALK, activin receptor-like kinase; ANG, angiopoietin; CKD, chronic kidney disease; DKD, diabetic kidney disease; ECM, extracellular matrix; Endo-MT, endothelial-mesenchymal transition; eNOS, endothelial nitric oxide synthase; EphB4, ephrin type-B receptor 4; Ephrin-B2,, erythropoietin-producing human hepatocellular receptor-interacting protein B2; FOXO1, forkhead box 01; ICAM1, intercellular adhesion molecule 1; JNK, c-Jun N-terminal kinase; LRG1, leucine-rich α-2-glycoprotein 1; NRP1, neuropilin 1; PEDF, pigment epithelium-derived factor; PEDFR, pigment epithelium-derived factor receptor; PTC, peritubular capillary; Smad, suppressor of mothers against decapentaple; sVEGFR1, soluble vascular endothelial growth factor receptor 1; TGF-β, transforming growth factor-β; TGF-βR, transforming growth factor-β receptor; TIE2, tyrosine-protein kinase receptor 2; TIF, tubulointerstitial fibrosis; VASH, vasohibin; VCAM1, vascular cell adhesion molecule 1; VEGF-A, vascular endothelial growth factor A; VEGFR2, vascular endothelial growth factor receptor 2; VEPTP, vascular endothelial protein tyrosine phosphatase; WNT, Wingless-type MMTV integration site.

The principal mouse models of kidney TIF, which have been employed in the studies referenced in this review, include the unilateral ureteral obstruction (UUO)^[93] and the unilateral ischemia-reperfusion injury (UIRI) models.^[94] The UUO model is widely accepted because it faithfully replicates many of the pathological features of TIF,^[93] including tubular injury and atrophy, interstitial infiltration of inflammatory cells, myofibroblast activation, and deposition of ECM, whilst UIRI is a very robust model to study the progression from AKI to long-term TIF.^[94] In addition, the streptozotocin (STZ)^[95] and db/db^[96] mouse models are established research tools for studying type-1 and type-2 diabetes mellitus (DM) respectively and are used to investigate the early features of diabetic kidney disease (DKD).

The three cardinal groups of angiogenesis-related factors that regulate the normal structure and function of the vascular niche are VEGF (including its associated receptors and modulators), angiopoietins and transforming growth factor beta (TGF-β) (Figure 3). VEGF-A is the major isoform of human VEGF secreted from podocytes and tubular epithelial cells.^[97,98] It activates the VEGF receptor-2 (VEGFR2) expressed on endothelial cells and promotes vascular sprouting in angiogenesis.^[99] VEGF-A also binds to a co-receptor, neuropilin-1 (NRP1), which is required for activating VEGFR2^[100] and triggering downstream signaling pathways. These include: i) phospholipase-Cγ (PLCγ)-protein kinase-C (PKC)-extracellular signal-regulated kinase (ERK) signaling inducing endothelial cell proliferation; ii) steroid receptor coactivator (SRC)-focal adhesion kinase (FAK) signaling leading to migration; iii) phosphatidylinositol 3-kinase (PI3K)/AKT pathway for endothelial cell survival; iv) SRC-induced internalization of vascular endothelial cadherin leading to increased vascular permeability.^[101] In contrast to VEGFR2, the activation of VEGFR1 results in much lower pro-angiogenic activity. Therefore, the circulating soluble version of VEGFR1 can serve as a VEGF-A antagonist.^[73] The role of podocyte VEGF-A in the glomerular microvasculature has already been demonstrated with VEGF-A knockout resulting in endothelial injury and glomerulosclerosis in adult transgenic mice with STZ-induced type-1 DM.^[102] Increased VEGF gene expression in podocytes also correlates with glomerular neovascularization in human DKD.^[103] Similarly, tubular VEGF-A has a pivotal role in maintaining the PTC network as evidenced by a significant reduction in PTC surface area in Vegfaflox/flox;Pax8-rtTA;Tet-O-Cre mice, in which VEGF-A is specifically excised in kidney tubular epithelial cells upon doxycycline administration.^[24] Conversely, increased tubule-specific VEGF-A production in bi-transgenic Pax8-rtTA/(tetO)7-VEGF mice causes an enlargement of PTCs with a higher density of endothelial cells.^[104]

Figure 3.

Major angiogenic signaling pathways in endothelial cells.

Certain angiocrine factors expressed in the vascular niche inhibit the VEGFA-VEGFR2 signaling pathway. These include semaphorin-3A, which is secreted by podocytes^[105] and binds to the NRP1 receptor,^[106] as well as pigment epithelium-derived factor (PEDF), which is secreted by endothelial cells^[77] and cleaves VEGFR2 by enhancing y-secretase activity at the transmembrane region.^[107] Podocyte-specific semaphorin-3A is elevated in human advanced DKD and semaphorin-3A overexpression promoted severe nodular glomerulosclerosis in mice with STZ-induced DM.^[108] PEDF expression was also found to be decreased in Sprague-Dawley rats with adriamycin-induced nephropathy, resulting in increased VEGF expression. Exogenous PEDF administration suppressed the elevated VEGF expression and subsequently reduced proteinuria in this model.^[109] Similarly, ephrin-B2 (erythropoietin-producing human hepatocellular receptor interacting protein), expressed in endothelial cells and pericytes,^[74] acts as a modulator of VEGFA-VEGFR2 signaling. Ephrin-B2, through its interaction with the ephrin type-B receptor-4 (EPHB4),^[110] controls the internalization and localization of VEGFR2, essentially fine-tuning VEGF signaling within endothelial cells for proper vascular remodeling and stability in the kidney. Ephrin-B2-EPHB4 signaling was found to play a vital role in angiogenesis, with increased PTC rarefaction demonstrated in UUO and UIRI mouse models with impaired PDZ-dependent ephrin-B2 signaling.^[75]

Angiopoietins (ANG), particularly ANG-1 and ANG-2, are a group of essential angiogenic factors involved in vessel homeostasis in the vascular niche through their interactions with the endothelial TIE-receptor tyrosine-kinases.^[81] ANG-1, expressed in tubular epithelial cells, stimulates its receptor TIE2 and mediates vascular stability and quiescence by eliciting an antiinflammatory response in endothelial cells through downregulation of VCAM1 and intercellular adhesion molecule-1 (ICAM-1). This downregulation is mediated via inhibition of the nuclear factor k-light-chain-enhancer of activated B-cells (NF-kB) pathway.^[79] Conversely, ANG-2, released from the Weibel-Palade bodies of endothelial cells into the circulation in response to external stimuli such as inflammation and ischemia/hypoxia,^[111] acts as a competitive antagonist of ANG-1 mediated TIE2 activation and promotes endothelial cell destabilization, migration and proliferation.^[81] In addition, vascular endothelial tyrosine phosphatase (VEPTP), expressed on endothelial cells, suppresses TIE2 activation by dephosphorylating its intracellular domain.^[82] ANG-2 modulates angiogenesis in the vascular niche in a VEGF-A/VEPTP dependent manner. It upregulates vessel sprouting, vascular permeability and inflammatory infiltration in the presence of elevated levels of VEGF-A and VEPTP, but exhibits anti-angiogenic properties with lower levels of VEGF-A and VEPTP in the microenvironment leading to vascular rarefaction.^[112,113] ANG-1 was found to be crucial in limiting the enhanced angiogenic and fibrogenic response in the presence of elevated levels of VEGF-A, ANG-2 and TGF-β, when comparing STZ-induced diabetic mice with and without ANG-1 deletion.^[80] In contrast, ANG-2 causes increased albuminuria and endothelial apoptosis in C57BL/6J mice with podocyte-specific ANG-2 overexpression.^[114] Elevated plasma ANG-2 levels and higher ANG-TIE pathway activity in endothelial cells have been predictive of disease progression in DKD patients.^[115]

More recently, vasohibins were characterized as a family of endothelium-derived regulators of angiogenesis. Vasohibin-1 (VASH-1) is secreted in areas of vessel sprouting in response to VEGF-A stimulation to counteract its pro-angiogenic effects^[116] by preventing endothelial proliferation and migration. VASH-1 also promotes endothelial survival through upregulation of sirtuin-1 (SIRT1) and superoxide dismutase-2 (SOD2).^[88] Vasohibin-2 (VASH-2), however, increases angiogenesis by inhibiting the termination of vessel sprouting.^[117] Given its role in promoting endothelial survival, VASH-1 heterozygous knockout (VASH-1+/−) mice with cisplatin-induced kidney injury exhibited increased PTC rarefaction.^[89] Similarly, VASH-2 deficient mice were also found to have markedly accelerated PTC loss in a UIRI model.^[90] Interestingly, clinical data suggest that serum VASH-1 is increased in type-2 DM patients,^[118] which likely underscores its role as a negative feedback regulator in response to endothelial injury and the elevated pro-angiogenic activity of VEGF in DKD.

In addition, TGF-β modulates various aspects of endothelial function including cell proliferation and apoptosis, and consequently, can propel both pro-angiogenic and anti-angiogenic pathways in the vascular niche.^[84] For instance, TGF-β induced activation of Smad2/Smad3 via activin receptor-like kinase-5 (ALK5) inhibits angiogenesis by downregulation of pro-angiogenic genes such as the kinase insert domain receptor (KDR) gene, which encodes VEGFR2.^[119] Conversely, TGF-β induced ALK1 activation stimulates the Smad1/Smad5/Smad8 pathway to promote endothelial proliferation, migration and angiogenesis.^[85] Leucine-rich alpha-2-glycoprotein-1 (LRG1), a regulatory factor that promotes TGF-β mediated pro-angiogenic activity, has been shown to contribute to DKD progression via potentiation of ALK1-Smad1/5/8 signaling in mice with STZ-induced DM.^[86] Elevated plasma LRG1 levels were also associated with unfavorable kidney outcomes in DKD patients.^[86]

Finally, the distinct physiological demands and microenvironments of the kidney cortex and medulla necessitate differences in angiocrine signaling pathways within their respective vascular niches. Single-cell RNA sequencing of mouse kidneys showed heterogeneous expression of various angiocrine factors among different endothelial cell phenotypes.^[42,43] For instance, identified markers of medullary endothelial cells included the Adiponectin Receptor-2 (ADIPOR2) and the 15-hydroxy prostaglandin dehydrogenase (HPGD) genes involved in vasodilation,^[42,43] contributing to medullary blood flow and osmolarity gradient regulation.^[120] In addition, IGFBP7 is highly expressed in medullary endothelial cells, whereas IGFBP3 and IGFBP5 is enriched in cortical endothelial cells.^[42,43] Further exploration is warranted of the precise mechanisms and interactions of the IGF signaling system, as well as several other angiocrine factors within the unique microenvironments of different kidney compartments.

The Role of the Vascular Niche in Regulating Kidney Tubulo-Interstitial Fibrosis and the Vascular Niche as a Potential Therapeutic Target in Kidney Disease

Although the emphasis has predominantly been on maladaptive tubular epithelial repair as a mechanism for driving kidney fibrosis,^[121] other cellular and molecular alterations in the structure and steady state of the vascular niche may also be important contributing factors. For example, dysregulated expression of angiogenic and angiostatic regulatory factors, capillary endothelial damage, separation of pericytes from endothelial cells, endothelial-mesenchymal transition (Endo-MT), as well as the transformation of pericytes and macrophages into myofibroblasts may all promote the development and progression of kidney TIF.^[122] Therefore, the vascular niche may be considered a potential target for anti-fibrotic therapies which focus specifically on interfering with the angiocrine function of endothelial cells, inhibiting Endo-MT, preventing the transformation of pericytes into myofibroblasts as well as regulating macrophage function within the vascular niche.

The imbalance between pro-angiogenic and anti-angiogenic factors in kidney disease leads to a cascade of events including inflammation, endothelial cell apoptosis and activation of myofibroblasts, culminating in the deposition of ECM components and fibrosis.^[123] The expression of VEGF-A in the vascular niche has been posited to influence TIF through the regulation of nitric oxide metabolism, endoplasmic reticulum stress and epithelial-mesenchymal transition.^[124] Overexpression of kidney VEGF-A in transgenic rabbits resulted in glomerular capillary proliferation and proteinuria, eventually culminating in the development of glomerular sclerosis with tubular atrophy and interstitial fibrosis.^[125] However, the exogenous administration of VEGF-A as a therapeutic strategy led to the preservation of PTC density and ameliorated kidney TIF in Sprague-Dawley rats with bilateral IRI.^[126] Given the available evidence, it is unclear whether VEGF-A potentiation or blockade has an effect on kidney fibrosis. On the other hand, ANG-2 expression not only enhances C-C-motif-ligand-2 (CCL2) mediated macrophage infiltration but also induces endothelial cell apoptosis in kidney TIF. As a result, recombinant L1–10, which is an ANG-2 inhibitor, was found to have inhibitory effects on macrophage infiltration, vascular rarefaction and fibrosis in both UUO and UIRI models of progressive kidney disease.^[28] Similarly, an agonistic TIE2 monoclonal antibody, mimicking the endogenous actions of ANG-1, suppressed TIF through the attenuation of inflammation and endothelial apoptosis in CD-1 mice with folic acid-induced nephropathy.^[127] Moreover, EPHB2, a component of the Eph-receptor tyrosine-kinase family involved in endothelial function and angiogenesis,^[128] was found to be significantly upregulated in type-2 diabetic db/db mice as well as in the disease-involved tubular areas in kidney biopsies of IgA nephropathy patients.^[29] Globally knocking out EPHB2 in C57BL/6J mice significantly protected their kidneys from UIRI-induced fibrosis and reduced several pro-fibrotic pathways, including inflammatory cytokine upregulation, tubular epithelial-mesenchymal transition and myofibroblast activation. Importantly, these EPHB2 knockout mice featured normal kidney structure and function under basal conditions, suggesting no major role for EPHB2 in kidney development.^[29] This suggests that the EPHB2-receptor tyrosine-kinase signaling pathway may be a viable target for the treatment of kidney TIF.

Endo-MT refers to endothelial cells in the vascular niche undergoing cellular trans-differentiation, losing their characteristic endothelial features and transforming into myofibroblasts in response to injury.^[129] As such, they contribute to the development of kidney fibrosis by producing ECM components and promoting tissue remodeling. Early lineage tracing experiments, using techniques like Tie2-Cre mice, labeled endothelial cells and found that a substantial portion of myofibroblasts expressed endothelial markers in STZ-induced diabetic mice and in UUO nephropathy.^[130,131] However, more recent studies with advanced fate-mapping techniques have challenged the notion of widespread Endo-MT, finding minimal contribution of Endo-MT to the myofibroblast pool, with pericytes and resident fibroblasts identified as the main precursors of myofibroblasts.^[132] Another study found that only approximately 10% of the myofibroblast population originated from endothelial cells via Endo-MT in UUO injury.^[133] Therefore, the existing evidence largely indicates that endothelial injury contributes to fibroblast differentiation and fibrosis primarily through paracrine mechanisms influencing neighboring cells, particularly pericytes. However, one newer endothelial lineage tracing study in a UUO mouse model detected evidence of both incomplete Endo-MT, where 13% of the labeled cells co-expressed the endothelial marker CD31, and the myofibroblast marker, α-smooth muscle actin (αSMA), as well as a more advanced form of Endo-MT, with 10% of the labeled cells expressing αSMA without detectable CD31.^[134] The role of TGF-β in Endo-MT has also been investigated, with TGF-β pathway inhibition, utilizing a specific inhibitor of Smad3 (SIS3), leading to abrogation of Endo-MT and the amelioration of fibrosis in STZ-induced DKD in Tie2-Cre;Loxp-EGFP mice.^[135] Moreover, the mineralocorticoid receptor blocker, esaxerenone, has been found to effectively inhibit kidney angiogenesis and Endo-MT by modulating the VEGF-A and TGF-β1 pathways in aldosterone-infused C57BL/6J mice.^[30] Similarly, the endothelial glucocorticoid receptor has been reported to play a pivotal regulatory role in the process of Endo-MT in DKD by downregulating the WNT signaling pathway and mesenchymal transition, thereby mitigating kidney fibrosis.^[136] Therefore, blockade of Endo-MT by exploiting the TGF-β, mineralocorticoid and glucocorticoid receptor signaling pathways may hold therapeutic potential in the context of kidney disease.

The detachment of pericytes from endothelial cells lining the capillaries may represent a critical step in the development of kidney fibrosis as it allows pericytes to migrate into the interstitial space and transform into scar-forming myofibroblasts.^[137] Targeting the restoration of pericyte-endothelial cell interaction may offer a potential therapeutic strategy in kidney disease, but this has not yet been demonstrated.^[138] Furthermore, even though genetic lineage tracing studies have provided evidence for pericytes as major progenitors of myofibroblasts in kidney fibrosis,^[139] there are conflicting data which suggest that pericytes, similar to endothelial cells and tubular epithelial cells, do not contribute significantly to the myofibroblast pool.^[133] Consequently, it remains to be determined whether pericytes in the vascular niche can be targeted to inhibit the emergence of myofibroblasts in kidney TIF.

Lastly, macrophages not only participate in immune surveillance and tissue homeostasis within the vascular niche, but recent evidence has also indicated that they can directly transdifferentiate into myofibroblasts in the process of MMT in response to kidney injury, leading to ECM deposition and fibrosis.^[51,70,140] Detecting MMT relies on identifying cells that co-express markers of both macrophages (F4/80 or CD68) and myofibroblasts (collagen I [Col-I] or αSMA), using methods such as immunofluorescence, flow cytometry, lineage tracing and single-cell RNA sequencing.^[51,140,141] For instance, MMT was found to be an important contributor to TIF in human and experimental chronic kidney allograft injury.^[51] The study utilized lineage tracing in Lyz2-Cre/Rosa26-Tomato C57BL/6J mice with BALB/c kidney allografts to demonstrate that 37% of the total αSMA+ myofibroblast population in the kidney allografts originated from recipient bone marrow-derived macrophages.^[51] Similarly, lineage tracing demonstrated that MMT cells accounted for 65%–70% of total Col-I+ or αSMA+ myofibroblasts in a UUO model, suggesting that the MMT process is influential in chronic inflammation-induced kidney fibrosis.^[140] Conversely, in another study involving the UUO experimental model, Kramann et al.,^[141] utilizing a combination of lineage tracing and single-cell RNA sequencing, found that circulating monocytes directly contribute only a minor fraction of myofibroblasts, and that their role in kidney fibrosis might be more indirect, primarily by promoting inflammation and paracrine activation of resident mesenchymal cells (pericytes and resident fibroblasts). Despite the conflicting evidence surrounding the MMT process in the kidney, potential anti-fibrosis agents for kidney disease have been tested with respect to their role in MMT in experimental TIF models, including the mineralocorticoid receptor antagonist, eplerenone,^[142] the P2Y12 inhibitor, clopidogrel,^[143] as well as a fatty acid-binding protein-4 (FABP4) inhibitor.^[144]

Conclusions and Future Directions

A growing body of evidence suggests that the microenvironment of the kidney vasculature, which may be referred to as the ‘vascular niche’, can modify the course of various kidney diseases through its influence on the development and progression of TIF.^[71] Under physiological conditions, the components of the vascular niche including capillary endothelial cells, tubular epithelial cells, pericytes and inflammatory cells maintain microvascular homeostasis by engaging in crosstalk with their neighboring cells via the balanced expression of angiocrine regulatory factors.^[37] However, in the presence of pathological stimuli such as capillary endothelial injury, the intricate balance of angiocrine factors in the vascular niche is disrupted, leading to a cascade of excessive inflammation, pathological repair and fibrosis.^[15] Consequently, the exploitation of the vascular niche may be a promising approach to ameliorate kidney TIF by developing therapeutic strategies targeting angiogenesis, the angiocrine function of endothelial cells, Endo-MT as well as those designed to suppress the transformation of pericytes and macrophages into myofibroblasts.

Although PTC rarefaction within the vascular niche has been considered a hallmark pathologic feature of kidney TIF, the mechanisms underlying this process remain incompletely understood.^[145] Conversely, there are several murine^[28–30] and human studies^[25–27] that have demonstrated that the upregulation of pro-angiogenic biomarkers and pathways is closely associated with the development of TIF and kidney disease. Although the cited studies^[25–30] highlight the potential role of angiogenic and inflammatory proteins in kidney TIF and CKD, not only are longitudinal studies needed to determine whether the biomarkers identified in CKD patients can predict CKD progression, but further mechanistic studies are also required to fully unravel their contributions to kidney disease progression and identify potential therapeutic targets. Nevertheless, these data suggest that PTC rarefaction may instead be a sequela of advanced tubulo-interstitial damage rather than an initiating factor that triggers the development of kidney TIF itself.^[17] In fact, one can hypothesize that the upregulation of angiogenesis, particularly in the earlier stages of kidney disease, contributes to the advancement of TIF and CKD progression by promoting immune cell recruitment and activation. Correlative insights regarding this hypothesis may be obtained by conducting morphometric analyses of kidney biopsy specimens, especially early in the disease process and well before overt and extensive fibrosis has occurred. Such studies, exploiting three-dimensional (3D) imaging techniques^[146] and other protein and gene expression-based spatial techniques, will facilitate the quantification of PTC density, assessing its correlation with the degree of kidney TIF, as well as characterizing the dynamic cellular and molecular microenvironments of the vascular niche. In addition, certain geometric and hemodynamic parameters that may be quantified using morphometric analyses of 3D imaging data such as vessel diameter, curvature, tortuosity as well as branching point angle may contribute to a dynamic biomechanical environment within the kidney’s vascular niche.^[147] These biomechanical alterations may significantly impact endothelial cell integrity and function and therefore, play a potential role in maintaining endothelial health and influencing kidney disease susceptibility and progression.^[148]

Moreover, vascular niches are continually being transformed by a combination of angiogenesis, remodeling and pruning in response to a plethora of metabolic and hemodynamic stimuli,^[64] potentially leading to the establishment of diverse unique vascular niches. Within these capillary networks, the ‘downstream’ ends of individual unconnected vessel sprouts or ‘dead-end’ vessels have been referred to as ‘terminal points’, whereas the sites along a single capillary where it divides into multiple smaller capillary segments have been referred to as ‘branch points.’^[149] These terminal points and branch points may be conceptualized as examples of different types of vascular niches in the kidney parenchyma and have not been systematically studied to date (Figure 4). Spatial protein and gene expression approaches may potentially elucidate and resolve the unique cellular and molecular environments and interactions that characterize these niches and thus, identify potential therapeutic targets as well as biomarkers of kidney TIF.^[150]

Figure 4.

Types of vascular niches: “terminal points” and “branch points.”