Targeting Metabolic Reprogramming in Polycystic Kidney Disease

The kidney is a wonderfully complex organ and one of the most energy-demanding of our body.¹ Accordingly, some segments of the renal tubule are particularly dense with mitochondria, the cellular organelles responsible for generating ATP. However, mitochondria are much more than powerhouses and perform various fundamental functions.² For example, during proliferation, the mitochondrial metabolic flow is diverted to favor the release of essential metabolites required for biomass synthesis.²

In other words, proliferating cells need to switch from a catabolic mode (i.e., break down molecules to generate energy) to an anabolic mode (i.e., build up molecules to generate a new cell).² Under these conditions, energy production is delegated to cytosolic glycolysis and other carbon sources (such as amino acids) used to generate TCA cycle intermediates (Figure 1).² Many of the metabolic processes observed in proliferating tissues have been better understood when studied in the context of cancer, the hyperproliferative pathology par excellence.²

Figure 1.

Warburg effect is a feature of ADPKD. Schematic representation of the use of glucose in differentiated tissues (left) as opposed to proliferating, cystic, or cancer tissues (right). Full oxidation of glucose in mitochondria provides abundant ATP, whereas its oxidation in the cytosol provides little amounts and only when oxygen is lacking differentiated cells use anaerobic glycolysis. In cells that are proliferating, as well as in cystic tissues and cancers, whether in the presence or absence of oxygen, cells use cytosolic glycolysis to generate ATP and use intermediates of the TCA cycle as building blocks. This process is called aerobic glycolysis to indicate that it occurs also in the presence of oxygen, or the Warburg effect after its discoverer. ADPKD, autosomal dominant polycystic kidney disease.

Autosomal dominant polycystic kidney disease (ADPKD) is the most frequent hereditary pathology seriously affecting human health.³ Most patients inherit a mutation in the PKD1 or PKD2 gene in heterozygosity.³ However, the second allele of the same gene mutates somatically in the renal tubule during an individual's lifetime, causing the formation of cystic, clonal structures (originating from a single cell) that grow in number and size throughout the individual's life.⁴ These growing cysts eventually compress and impair the healthy parenchyma of the kidney, leading to declining function. This two-hit mechanism molecularly likens ADPKD to a tumor, albeit a very benign one, corroborating the earlier description of polycystic kidney disease (PKD) as a “cancer in disguise.”^4,5

It is therefore only partially surprising the discovery made by my group approximately 10 years ago that ADPKD shares profound metabolic reprogramming with cancer. In fact, we demonstrated that PKD is characterized by a glycolytic switch where cystic epithelia use large amounts of glucose, producing energy almost exclusively through aerobic glycolysis rather than by fueling the TCA cycle to produce ATP (Figure 1).⁶ This effect, which occurs in almost all tumors, is called the Warburg effect because it was first described by the German biochemist Otto Warburg in cancer cells.² Today, we know that this effect is quite common not only in cancer cells but also in normal proliferating cells (Figure 1).² On the basis of this metabolic switch, we described that a glucose analog called 2-deoxy-D-glucose (2DG) is particularly effective in slowing down PKD in various murine models of the disease (Figure 2).^6,7

Figure 2.

Metabolic rewiring in PKD and targets under development. Cystic epithelia show a very complex metabolic rewiring. Here, a simplified scheme of major central carbon metabolism rewiring is illustrated. Pyruvate is converted into lactate to generate energy and only minimally fueled into mitochondria. Glutamine utilization, which in PKD is driven by the enzyme ASNS, compensates for the lack of pyruvate import by fueling glutamate into mitochondria. Glutamine anaplerosis maintains mitochondria viability, while reductive carboxylation is used to synthesize large amounts of citrate, exported to the cytosol and converted to acetyl-CoA to fuel de novo fatty acid biosynthesis. Coordination of these alterations sustains proliferation and survival. 2DG counteracts glycolysis, while ASNSi blocks glutaminolysis. 2DG, 2-deoxy-D-glucose; ASNS, asparagine synthetase; ASNSi, inhibition of asparagine synthetase; PKD, polycystic kidney disease.

Studies by other groups confirmed that there is a very robust efficacy when 2DG is administered to rat or mini-pig models of PKD. The robust efficacy that we observed in animal models prompted us to design a strategy to test its possible efficacy on patients with ADPKD. There is extensive prior literature on the use of 2DG in humans, including trials on patients affected by cancer.⁸ The compound is generally well tolerated in humans,⁸ and pharmacokinetic studies conducted in animals have led us to conclude that the effective dose we use in mice (100 mg/kg) translates to a very well-tolerated dose on the basis of past human studies (human equivalent dose approximately 20 mg/kg) (Chiaravalli et al., in preparation). We are now in the process of completing all investigational new drug application–enabling studies required for initiating a first safety trial in patients affected by ADPKD.

2DG is a very old molecule never patented, so we hold an IP protection in the form of method of use that we think might ensure interest from private entities to complete the entire regulatory path impossible to pursue in academia. One key question that might arise: 2DG was abandoned in cancer not on the basis of safety or tolerability issues, but on the basis of lack of efficacy.⁸ Why should this compound now be so effective in ADPKD? Because, as said above, ADPKD shares a few features with cancer (Figure 1), but it is not a cancer. Growth of the cystic lesions is extremely slow, but continuous over time. It takes 50 years on average for innumerable cysts to grow and compromise kidney function.³ We believe that this relatively slow progression of growth in the cysts is what provides a unique window of opportunity for intervention in this disease.

Our data above on the complete dependency of cysts on glucose for energy production (Figure 1), however, also call for the question as to how mitochondria, mostly deprived of pyruvate import to fuel the TCA cycle, cope with this stress maintaining the mitochondrial membrane potential to avoid collapse and cell death. Here, again, cancer and previous studies on proliferating cells came to help. The literature shows that mammalian cells are flexible to use different sources of carbon for fueling the TCA cycle.² Indeed, in most cancers, an enhanced utilization of glutamine compensates for the lack of pyruvate import into mitochondria.² Similarly, in a thorough metabolic tracing study using isotopolog labeling, we discovered that cells lacking the Pkd1 gene use abundant amounts of glutamine.⁹ This glutamine is converted to glutamate and imported in mitochondria where a small amount is used in the oxidative side of the TCA cycle, whereas a larger amount is used in a reductive carboxylation process, which generates large amounts of citrate, exported to the cytosol, broken down to generate acetyl-CoA, next incorporated into newly synthesized fatty acids (Figure 2).⁹

Thus, preventing glutamine utilization blocks proliferation and survival in Pkd1 mutant cells; a synergistic effect is achieved when both glucose and glutamine utilization are blocked.⁹ Of great interest, during our isotopolog-labeling experiments with ¹⁵N₂-glutamine, we serendipitously discovered that the utilization of glutamine in PKD is driven by the enzyme asparagine synthetase (ASNS).⁹ This identified an appealing target for therapy because ASNS is expressed at extremely low levels in adult tissues and it only becomes expressed under stress conditions in adults, and it is upregulated in PKD tissues.^9,10

On the basis of this, we designed antisense oligonucleotides (ASOs) directed against murine Asns and followed PKD progression upon their administration. Asns-ASO had a great effect on retarding disease progression in murine models of PKD, including reduced kidney volume, cystic index, kidney/body weight, and, importantly, restored kidney function almost to completion.¹⁰ Thus, ASNS represents a novel target for therapy in PKD.¹⁰ Furthermore, combining 2DG with Asns-ASO results in an additive efficacy in halting PKD progression in animal models. We have filed a patent on inhibition of ASNS in PKD using ASO or small molecules as a therapeutic strategy.¹⁰

In conclusion, our work has uncovered a profound metabolic reprograming in PKD, limiting flexibility into using different sources of carbon. As such, the glycolysis inhibitor 2DG holds a great promise as a possible therapy for PKD. Our data also have uncovered some uniqueness in the rewiring of glutamine in PKD, which depends on the enzyme ASNS, in itself targetable for therapy. Combination of the two approaches further limits the ways of escape to the diseased cells and tissues leading to improved efficacy. While these initial targets are being developed and tested in clinical trials, future studies should concentrate on identification of the metabolic bypasses used by cells when exposed to 2DG or inhibition of ASNS as to further reduce flexibility and crack the system of energy production and cell survival in PKD.

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/JSN/E844.

Funding

A. Boletta: Ministero della Salute (RF-2018-12368254 and GR-2016-02364851), Congressionally Directed Medical Research Programs (PR210615), and Associazione Italiana Rene Policistico.

Author Contributions

Conceptualization: Alessandra Boletta.

Writing – original draft: Alessandra Boletta.