Wang and Luo Exp Hematol Oncol (2021) 10:30 https://doi.org/10.1186/s40164-021-00223-4 Experimental Hematology & Oncology

REVIEW Open Access The metabolic adaptation mechanism of metastatic organotropism Chao Wang1 and Daya Luo2*

Abstract Metastasis is a complex multistep cascade of cancer cell extravasation and invasion, in which metabolism plays an important role. Recently, a metabolic adaptation mechanism of cancer metastasis has been proposed as an emerg- ing model of the interaction between cancer cells and the host microenvironment, revealing a deep and extensive relationship between cancer metabolism and cancer metastasis. However, research on how the host microenviron- ment afects cancer metabolism is mostly limited to the impact of the local tumour microenvironment at the primary site. There are few studies on how diferences between the primary and secondary microenvironments promote metabolic changes during cancer progression or how secondary microenvironments afect cancer cell metastasis preference. Hence, we discuss how cancer cells adapt to and colonize in the metabolic microenvironments of difer- ent metastatic sites to establish a metastatic organotropism phenotype. The mechanism is expected to accelerate the research of cancer metabolism in the secondary microenvironment, and provides theoretical support for the genera- tion of innovative therapeutic targets for clinical metastatic diseases. Keywords: Cancer metastasis, Cancer metabolism, Organotropism, “Seed and soil” hypothesis

Background of micrometastatic clones and colonization in distant Cancer metastasis organs to form clinically detectable macrometastatic Metastasis refers to the process by which cancer cells clones [2, 3]. leave the site of cancer occurrence, arrive at distant In 1889, the English surgeon Stephen Paget discovered organs by various ways, and thrive in distant organs, that the organs of cancer metastasis are not randomly but which is one of the hallmarks of cancer cells [1]. Clini- selectively distributed; therefore, he proposed the “seed cally, approximately 90% of cancer patients die from and soil” hypothesis in which cancer cells are referred to metastatic disease, yet the molecular mechanisms under- as seeds and the destination of metastasis as represented lying metastasis are poorly understood [2]. It is currently by soil. Cancer cells can spread similar to seeds but can believed that cancer metastasis is a dynamic and com- thrive only in congenial soil [4]. Now, this phenom- plex multistep process known as the “metastatic cas- enon of non-random metastasis is known as “metastatic cade”. Tis cascade includes local invasion of surrounding organotropism”, or “organ-specifc metastasis”. Te com- tissues, intravasation to the blood vessel wall, survival mon metastatic organs vary by cancer type and subtype. in the circulatory system and formation of circulating Prostate cancer often metastasizes to bone [5], while tumour cells (CTCs), arrest in a distant organ, extrava- pancreatic cancer and uveal melanoma often metasta- sation as disseminated tumour cells (DTCs), formation size to the liver [6, 7]. Breast cancer often metastasizes to multiple organs, such as bone, lung, liver, brain, and *Correspondence: [email protected] distant lymph nodes, but diferent subtypes of breast 2 Department of Biochemistry and Molecular Biology, School of Basic cancer have diferent propensities for metastatic sites [8– Medical Sciences, Nanchang University, Nanchang 330006, China 10]. In addition, cancer cells can also exhibit an ordered Full list of author information is available at the end of the article and hierarchical organ-specifc metastatic pattern. For

© The Author(s) 2021. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat​ iveco​ ​ mmons.org/​ licen​ ses/​ by/4.​ 0/​ . The Creative Commons Public Domain Dedication waiver (http://creat​ iveco​ mmons.​ org/​ publi​ cdoma​ in/​ ​ zero/1.0/​ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 2 of 16

example, colorectal cancer usually frst metastasizes to phosphate synthase (CPS) 1 increases; this change leads the liver and then to the lung [11]. Tere are many fac- to an increase in the CP, which moves from mitochondria tors that determine metastatic organotropism, including to the cytoplasm. Subsequently, the ammonia-derived CP circulatory system patterns, internal characteristics of enters the pyrimidine synthesis pathway as a CPS2 sub- the cancer, the tissue- or organ-specifc microenviron- strate to realize waste recycling and support rapid can- ment, and the interaction between cancer cells and the cer DNA synthesis and fast cell proliferation [24]. One host microenvironment [3, 4, 12, 13]. Recently, the met- study discovered "urea cycle disorders" in a wide range of abolic adaptation mechanism of cancer metastasis has cancers. Tese disorders involve the transfer of nitrogen been proposed as an emerging model of the interaction from the urea cycle; the activation of the multifunctional between cancer cells and the host microenvironment enzyme CAD protein composed of CPS2, aspartate car- [14], revealing a deeper internal connection between bamoyl transferase and dihydroorotase; and the increase cancer metabolism and metastasis. in pyrimidine synthesis [25]. In addition, a recent study found that p53 can transcriptionally inhibit the metabolic Cancer metabolism enzymes CPS1, ornithine transcarbamoylase and argin- Metabolic reprogramming is one of the important fea- ase 1 in the urea cycle, thereby inhibiting urea synthesis tures in the development and progression of cancer, and ammonia clearance. Accumulation of ammonia can and it is crucial for the survival and growth of cancer signifcantly downregulate the mRNA translation of the cells [15]. A century ago, German biochemist Otto War- rate-limiting enzyme ornithine decarboxylase in polyam- burg frst observed that cancer cells undergo metabolic ine biosynthesis, thus inhibiting polyamine biosynthesis changes. In the case of sufcient oxygen, the anaerobic and cell proliferation [22]. In summary, cancer cells have oxidation of cancer cells is enhanced, oxidative phospho- extremely high metabolic fexibility and can generate rylation (OXPHOS) is inhibited, and a large amount of energy, support the integrity of the cell membrane struc- glucose produces lactate through glycolysis in a process ture, synthesize biological macromolecules and regulate known as the “Warburg efect” or “aerobic glycolysis” gene expression. Tese biological processes promote the [16]. With the continued in-depth research, more meta- occurrence, survival and development of cancer. bolic characteristics of cancer cells have been gradually Similar to metastasis, cancer metabolism is not a static revealed [17], and cancer metabolism shows incredible and uniform process but a dynamic and heterogeneous fexibility and plasticity. Ketone bodies are mainly pro- process that is mainly determined by the inherent char- duced by the liver but cannot be consumed by normal acteristics of cancer cells and the external environment adult hepatocytes [18]. However, in the absence of nutri- [15, 26, 27]. Te inherent characteristics of cancer cells ents, hepatocellular cancer cells can induce the expres- mainly include oncogenic mutations and the tissue of ori- sion of OXCT1, a key enzyme in ketone catabolism, by gin. Oncogene and tumour suppressor gene mutations activating the mTORC2-Akt-SP1 signalling axis, thereby can regulate the metabolic phenotype of cancer through utilizing ketone bodies to generate energy and support oncogenic signalling pathways. Te PI3K-AKT signal- cancer progression [19]. Another study also found there ling pathway undergoes recurring mutations in a variety are alternative fatty acid desaturation pathways activated of cancers, which enables cells to take up an increased in liver and lung cancer cells [20]. Te urea cycle is the number of nutrients and synthesize biological macromol- main pathway by which ammonia, a metabolic waste, is ecules, even in cases of low levels of extracellular growth eliminated from the body. On the one hand, urea cycle factor [28]. Te inactivating mutations of oncogenes RAS initiators and intermediates can be transferred to other and MYC can also enhance the uptake of glucose and glu- metabolic pathways for anabolism [21]; on the other tamine, increase the biosynthesis of biological macromol- hand, they can also activate metabolism-sensing signal- ecules and maintain energy and redox balance through ling pathways and exert biological efects distinct from diferent mechanisms [29–31]. P53 is a common tumour the classical metabolic pathways, demonstrating the suppressor gene that is mutated or deleted in approxi- strong waste-recycling capacity of cancer cells [22]. In mately 50% of human cancers. P53 can inhibit glyco- breast cancer, ammonia can be recycled as a nitrogen lysis, enhance OXPHOS and resist oxidative stress [32]. source by glutamate dehydrogenase and then used in the Mutations in metabolic enzymes can also directly regu- synthesis of other amino acids, such as proline and aspar- late metabolic phenotypes. Multiple metabolic enzymes tate, to support the growing breast cancer biomass [23]. that catalyse the TCA cycle can be mutated; for example, Normally, ammonia-derived CP in mitochondria cannot IDH1 and IDH2 mutations can lead to the accumulation enter the cytoplasm for pyrimidine synthesis. However, of D-2 hydroxyglutarate [33, 34]. SDH and FH muta- in non-small cell lung cancer (NSCLC) with KRAS muta- tions lead to the accumulation of succinate and fumarate, tion and LKB1 deletion, the expression of carbamoyl respectively [35]. D-2 hydroxyglutarate, succinate and Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 3 of 16

fumarate can afect epigenetic modifcations and HIF-1α 5-fuorouracil [41]; however, the concentration of uric degradation and participate in the development of cancer acid in mouse plasma is one-tenth that of human plasma, [17]. and therefore the mouse plasma environment shows lit- Te tissue of origin also afects the metabolic charac- tle efect on UMPS activity or 5-fuorouracil-induced teristics of the cancer. Trough pan-cancer analysis of resistance of cancer cells [41]. Subsequently, cells of the metabolic gene expression profles, the metabolic gene same primary cancer located in diferent environments expression programme of primary cancer was found show metabolic diferences. As analysed by preopera- to be very similar to that in the normal corresponding tive multimodality imaging combined with intraop- source tissues [36, 37]. Moreover, tumours from difer- erative 13C-glucose infusions, diferent areas of NSCLC ent tissue origins driven by the same oncogene may have perfusion showed metabolic heterogeneity [42]. Finally, diferent metabolic patterns. For example, although both diferent cancers in the same environment have similar are MYC-induced mouse tumours, MYC-induced liver metabolic patterns. Both gliomas and brain metasta- tumours show reduced glutamine anabolism, but MYC- ses can be fuelled by oxidized glucose and acetate with- induced lung tumours show increased glutamine anabo- out glutamine oxidation [43]. Terefore, the efect of the lism [38]. Kras activation and Trp53 deletion in mouse environment on cancer metabolism can have an even NSCLC and pancreatic cancer show diferent types of greater efect than the intrinsic characteristics of the can- branched chain amino acid metabolism [39]. cer cells. In addition to the inherent characteristics of cancer However, research on how the host microenviron- cells, the external environment also has an important ment afects cancer metabolism is mostly limited to the impact on metabolism, and cancer metabolism exhibits impact of the local tumour microenvironment at the pri- an environmental dependence (Fig. 1). First, diferences mary site. Few studies have explored how the diference between the in vivo and in vitro environments lead to between the primary and secondary microenvironments diferences in cancer metabolism. Ras-driven NSCLC promotes metabolic changes during cancer progression mainly relies on glucose to supply the TCA cycle in the or how diferent secondary microenvironments shape body and mainly relies on glutamine to supply the TCA diferent cancer metabolic patterns. Hence, we discuss cycle in medium [40]. Second, environmental difer- how cancer cells adapt to and settle in the metabolic ences between animals and humans also lead to difer- microenvironment of diferent metastatic sites to estab- ences in cancer metabolism. Uric acid can directly inhibit lish a metastatic organotropism phenotype. Te results uridine monophosphate synthase (UMPS) and reduce are expected to accelerate the research of cancer metabo- the sensitivity of cancer cells to the chemotherapy drug lism in the secondary microenvironment and promote the creation of new therapeutic targets for clinical meta- static diseases.

PI3K signaling Oncogenic pathway; RAS; Metabolic adaptation mechanism of metastatic mutation MYC; TP53; organotropism IDH1/2; SDH; FH Te metabolic adaptation of metastatic organotropism refers to the fact that metabolic pattern of metastatic Cancer cancer cells needs to be compatible with the metabolic metabolism microenvironment of distant organs to support the sur- vival and growth of metastasizing cancer cells. Te meta- bolic microenvironment that drives cancer metastases to Tissue origin Environment distant organs needs to adapt to acquire sufcient energy, nutrient sources, organ-specifc metabolites, and the Glutamine and Differences between in vivo proper pH, adjust to hypoxic or normoxic conditions, and branched chain and in vitro and between support metabolic interactions between organ-specifc amino acid humans and animals; metabolism intratumor heterogeneity cells and cancer cells. Only a small proportion of DTCs can colonize at a distant site, indicating that the survival Fig. 1 Determinants of cancer metabolism. The inherent characteristics of cancer cells and the external environment together and growth of DTCs during colonization requires over- determine cancer metabolism. The inherent characteristics of cancer coming signifcant obstacles. Organ-specifc metabolic cells mainly include oncogenic mutations and tissue origin. However, adaptation can help metastatic cancer cells overcome the efect of the environment on cancer metabolism can be greater the obstacles faced when colonizing distant organs and than that of the intrinsic characteristics of the cancer cells establishing an organotropism phenotype [14]. Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 4 of 16

Bone provide markers for early clinical diagnosis and treat- Bone has a high mineral content, extreme hardness, high ment. In view of the presence of HA in the microcal- extracellular calcium concentration, a hypoxic microen- cifcation area and bone metastases in breast cancer, vironment and acidic pH [44, 45]. Te inorganic phase of predictions on the tendency of bone metastasis based on bone is mainly composed of mineral hydroxyapatite (HA) the existence and morphological structure of microcalci- crystals, while the bone matrix is rich in type I collagen, fcation in the primary site remain to be explored. Pros- osteopontin and bone sialoprotein [45]. In addition, the tate cancer, breast cancer, and renal cell carcinoma with a bone matrix is also rich in a variety of cytokines and tendency to metastasize to bone can sense the extracellu- growth factors [45]. Te bone microenvironment con- lar ­Ca2+ concentration through calcium-sensitive recep- tains many types of cells, such as osteoblasts, osteoclasts, tors to promote cell proliferation and migration [50–52]. haematopoietic stem cells, adipocytes and immune cells Terefore, whether metastatic cancer cells can increase [46]. Under normal conditions, osteoblasts and osteo- the extracellular ­Ca2+ concentration through phagocyto- clasts establish a dynamic balance between bone for- sis and degradation to exert biological efects remains to mation and decomposition [46]. Bone metastasis often be determined. occurs in prostate cancer, breast cancer and lung cancer. In the primary site, hypoxia is an important feature of Among these processes, bone metastasis of prostate can- the tumour microenvironment [53]. Similarly, the bone cer usually presents an osteoblastic metastasis pheno- metastatic microenvironment is also a hypoxic environ- type, while breast cancer and lung cancer often present ment [44]. Hypoxia and HIF-1 expression can promote an osteolytic metastasis phenotype [46]. Metastatic cells breast cancer bone metastasis and induce osteolytic interact with osteoblasts and osteoclasts in the bone metastasis by inhibiting osteoblast diferentiation and microenvironment to promote the release of growth fac- promoting osteoclast production [54]. Moreover, ER- tors in the bone matrix and thus promote the growth of negative breast cancer cells with hypoxia at the primary the cancer cells at the site of the metastasis, forming a site can secrete lysyl oxidase to promote the formation “vicious cycle” of bone metastasis [46]. of premetastatic niches in osteolytic bone [55], revealing Te metabolic adaptation of bone metastatic cells is the close relationship between hypoxia and bone metas- mainly related to the specifc metabolites, oxygen con- tasis. Hypoxia in the primary site can lead to the produc- tent and various cell types in the bone microenviron- tion of a large amount of lactic acid, causing a reduction ment. In mouse models of breast cancer, cancer cells in extracellular pH [53]. Te bone metastatic microenvi- are more likely to metastasize to sites containing imma- ronment can also induce the formation of an extracellular ture HA, such as smaller, incomplete, and non-oriented acidic environment due to the combined actions of oste- crystallized HA sites [47]. Breast microcalcifcations oclasts and bone metastatic cancer cells [56]. Terefore, are common in the primary site of breast cancer. Tere- studying the similarity of metabolic microenvironments fore, radiography of breast microcalcifcations is com- of the primary site and the metastatic site is helpful for monly used in clinical screening for breast cancer. Breast exploring the mechanism of organotropism. microcalcifcations can be divided into two types. Type In addition, the specifc cellular metabolic interactions I microcalcifcations are mainly composed of calcium in the bone metastatic microenvironment are also related oxalate crystals, which are common in benign tumours to bone-specifc metastasis. Serine metabolism is very [48]. Type II microcalcifcations are mainly composed important for osteoclastogenesis [57]. Compared with of HA and are common in benign or malignant tumours breast cancer cells with weak bone metastasis activity, the [48]. Microcalcifcation in the primary site of breast can- expression of de novo serine synthesis enzymes, includ- cer is related to cell proliferation and migration, infam- ing PHGDH, PSAT1, and PSPH, is increased in breast mation and matrix degradation [48]. Breast cancer cells cancer cells with high bone metastasis activity [58]. Ser- may increase the intracellular calcium ion concentration ine can stimulate the formation of human osteoclasts and through phagocytosis and the degradation of intracellular promote their osteolytic activity [58]. Terefore, breast crystals, thereby exerting cancer-promoting efects [49]. cancer cells may promote the formation of osteoclasts In addition, previous studies have shown that there is a through the de novo synthesis of serine to promote a pre-selection mechanism for cancer metastasis. Because vicious cycle. Te lactate released by MDA-MB-231 cells of similarities between the primary microenvironment through glycolysis can be taken up by osteoclasts to pro- and the secondary microenvironment, the propensity for vide energy, enhance their ability to degrade bone matrix, bone metastasis may pre-exist in the primary cancer [13]. and promote the formation of osteolytic metastasis [59]. Te existence of the pre-selection mechanism means Osteogenic cells can deliver Ca­ 2+ to cancer cells through that the primary tumour microenvironment can be used gap junctions and activate downstream calcium signal- to predict the organ specifcity of cancer metastasis and ling pathways and mTOR signalling pathways to support Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 5 of 16

the progression of bone metastasis [60]. In addition to blood supply system of the liver and the high permeabil- osteoclasts and osteogenic cells, the metabolic interac- ity of blood sinuses may provide appropriate secondary tions between bone marrow adipocytes and cancer cells sites. Second, due to the natural tolerance of the liver to can also promote cancer cells adaptation to the bone substances derived from intestinal microorganisms, an microenvironment. Bone metastatic prostate cancer cells immune-tolerant environment is developed [68]. can take up lipids from bone marrow adipocytes through Te metabolic adaptation of liver-specifc metastatic fatty acid-binding protein (FABP) 4 to promote cancer cells is mainly related to competition for hepatic nutri- growth and invasion [61–63]. Bone marrow adipocytes ents, the hypoxic environment, and the efects of intra- can also promote the activation of HIF-1α in metastatic hepatic cells. Metastatic liver cancer cells may compete prostate cancer cells, resulting in metastatic cells acqui- with liver cells for nutrients needed for growth by acquir- sition of the Warburg efect phenotype [64]. In addi- ing metabolic capacity similar to that of health liver cells. tion, primary cancer cells can also secrete exosomes to Low-dose fructose is mainly metabolized by the small deliver pyruvate kinase M2 to bone marrow stromal cells intestine, and high-dose fructose is mainly metabolized (BMSCs) and upregulate CXCL12 of BMSCs in a HIF- by colon microorganisms and the liver [69]. However, 1α-dependent manner, forming a premetastatic niche for liver metastatic colon cancer cells can upregulate aldolase bone metastasis of prostate cancer [65]. B via GATA6, thereby acquiring the ability to metabolize In summary, the current studies on the metabolic adap- fructose and enhancing the proliferation ability of colon tation of bone metastasis mainly focus on the roles of cancer cells [70]. A recent study showed that liver metas- inorganic components and bone-specifc cells. However, tasis of colorectal cancer can induce epigenetic remod- most of the efects of bone inorganic components on elling through enhancers or super enhancers, thereby cancer has been characterized based on in vitro tests, and causing the migrating colorectal cancer cells to lose more in vivo tests are needed to prove that these efects their colorectum-specifc transcription programme and truly occur. Te understanding of the efects of bone- acquire a liver-specifc transcription programme [71]. In specifc cells on cancer metabolism is still limited to the addition, this study also used the gene expression profles interaction between cancer cells and other cells. Tus, of a large number of primary and metastatic cancers to more studies are needed to demonstrate how multiple show the universality of transcription characteristic elim- cellular synergies afect the metabolism of bone metasta- ination from primary cancer cells and transcription char- ses. Because osteolytic metastasis can degrade not only acteristics acquisition of secondary site cells [71]. More the inorganic component HA in the bone matrix but also studies are needed to prove whether other types of meta- the organic component, studies should be taken to deter- static cancer cells also acquire the transcriptional and mine whether osteolytic metastasis cells can directly metabolic characteristics of the metastatic site through obtain organic nutrients from the matrix by promoting epigenetic remodelling. osteoclast function. Tere is a hypoxic microenvironment in the liver that can easily lead to the depletion of ATP that causes Liver metastatic cancer cells to die. Metastatic cancer cells in Te liver is the metabolic centre, where liver cells com- the liver from colorectal cancer can downregulate miR- plete most metabolic activities of the whole body to 483-5p and miR-551a to upregulate the expression of control substance and energy balances. Te liver has a a brain-type creatine kinase and secrete it into the liver unique blood supply system composed of the hepatic metastatic microenvironment, phosphorylating creatine artery and portal vein, which merge through the sinu- to produce phosphocreatine [72]. Subsequently, phos- soids and fnally fow into the central vein [66]. Arterial phocreatine is taken up by the metastasized colorectal blood from the hepatic artery is enriched with nutrients cancer cells to supply intracellular ATP, allowing them and oxygen, while venous blood from the portal vein con- to survive under hypoxic conditions [72]. A study on the tains food-derived substances such as lipid droplets and metabolic reprogramming of multiple organ metastases is deoxygenated [66]. Terefore, the liver has a hypoxic of breast cancer found that compared with breast cancer microenvironment. In addition, the liver microenviron- at the primary site, breast cancer cells that have metasta- ment is composed of a variety of cells, including hepato- sized to the liver, bone or lung show increased glycolysis cytes, hepatic stellate cells (HSCs), sinusoidal endothelial and tricarboxylic acid cycle activity levels [73]. However, cells and many kinds of immune cells [67]. Liver metas- compared with that of bone and lung metastasis, the tasis often forms with breast cancer, melanoma and gas- expression of HIF-1α and HIF-1α target PDK1 expres- trointestinal cancer cells. Te reason that the incidence of sion are at a higher level in liver cancer metastasis, result- metastatic liver cancer is higher than that of primary liver ing in stronger downstream glycolytic activity; however, cancer requires a two-part explanation. First, the special oxidation phosphorylation and glutamine metabolism Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 6 of 16

are weaker in the liver, which promotes the adaptation of cancer cells need to overcome the capillary barrier to col- cancer cells to the hypoxic environment in the liver [73]. onize. Tere are many types of cells in the lung, including Terefore, the hypoxic microenvironment in the liver type I alveolar epithelial cells, type II alveolar epithelial forces metastatic cancer cells to undergo greater energy cells, endothelial cells, fbroblasts, and various types of metabolism and enhances the characteristics of the War- immune cells [75]. Furthermore, in addition to its respi- burg efect. ration function, the lung also has many metabolic func- Intrahepatic cells can also change the metabolic char- tions. While the lung plays a prominent role in lipid acteristics of liver metastases. Under normal physiologi- metabolism and the synthesis of pulmonary surfactant; cal conditions, HSCs are in a resting state. However, on the other hand, it also synthesizes a variety of vaso- upon triggered infammation, HSCs are activated to active substances, including amines, prostaglandins and become hepatic myofbroblasts (HMFs). Te expression peptides [76]. level of succinate dehydrogenase B (SDHB) is higher and Currently, the metabolic adaptation of lung-specifc the OXPHOS activity is stronger when pancreatic ductal metastasis mainly focuses on mitochondrial metabo- epithelial cells (PDECs) were cocultured with HSCs than lism, antioxidant programmes, and the role of some cocultured with HMFs. [74]. Knocking down SDHB can metabolites or metabolic enzymes. Due to the oxygen- enhance cell proliferation and increase CSC character- rich microenvironment in the lungs, metastasizing can- istics [74]. Consistently, liver micrometastases (repre- cer cells can rewire their metabolism from glycolysis sentative of a non-infammatory environment) in KPC to OXPHOS. A recent study used a breast cancer PDX mice with pancreatic cancer exhibited higher expression model and single cell transcriptome sequencing technol- of SDHB and lower expression of the CSC marker nes- ogy to prove that lung and lymph node micrometastases tin than liver macrometastases (representative of a highly exhibit high levels of OXPHOS activity [77]. In contrast, infammatory environment) [74]. Tis fnding shows that primary breast cancer cells show upregulated glycolytic the activation level of HSCs and the infammatory state of activity [77]. Tis conclusion was also verifed by fow the liver can afect the metabolic state, proliferation abil- cytometry and metabolomics [77]. Mitochondrial com- ity and stem cell characteristics of PDECs. More evidence plex V inhibitor oligomycin can be used to inhibit breast is needed to prove that other types of cells in the liver can cancer OXPHOS and lung metastasis efectively but has afect the metabolic characteristics of liver metastases. no efect on primary breast cancer [77], indicating that In summary, liver metastatic cells can acquire the OXPHOS has a profound targeting efect on lung metas- specifc metabolic ability of the liver and adapt to the tasis, which also suggests that drugs targeting the mito- hypoxic environment. Te metabolic ability of liver chondrial electron transport chain can be used to target metastatic cells can also be afected by other cells in the lung metastases. Peroxisome proliferation-activated liver. In fact, there are other specifc metabolic features receptor coactivator 1 alpha (PGC-1α) is a core regula- of the liver, such as ketone body metabolism, enterohe- tor of mitochondrial energy production and oxidative patic circulation of bile acid, and ammonia metabolism, metabolism, and its expression is often altered during that may change the metabolic behaviour of cancer cells. cancer metastasis [78]. However, the specifc mechanism For example, the aforementioned liver cancer cells can of PGC-1α in cancer metastasis remains controversial. decompose ketone bodies to obtain substrates for energy A study found that CTCs can increase the rate of mito- metabolism [19]. Terefore, more studies are needed to chondrial OXPHOS, biosynthesis and oxygen consump- demonstrate that other metabolic characteristics of liver tion by upregulating PGC-1α to promote subsequent cells can infuence the metabolic behaviour of cancer lung metastasis [79]. Silencing PGC-1α slows lung metas- cells that have metastasized to the liver. tasis but does not afect primary cancer growth or the epithelial-mesenchymal transition (EMT) [79]. However, Lung another study found no signifcant efect of the OXPHOS Lung metastasis often follows breast cancer, melanoma, inhibitor metformin on metastasis in mice injected with and colon cancer [75]. Te lung is the respiratory organ breast cancer cells that highly express PGC-1α [80]. In through which oxygen is inhaled and carbon dioxide is contrast to the mechanism by which PGC-1α upregulates exhaled. Terefore, in contrast to the hypoxic environ- OXPHOS, this study indicated that PGC-1α enhances ment of the bone and liver, the lung contains high oxy- global energy metabolism fexibility of breast cancer cells, gen levels, and cancer cells need to overcome the stress thereby producing resistance to bioenergy interference caused by oxidative damage to colonize in the lung. Te drugs such as metformin [80]. Terefore, the relationship abundant capillary network in the lung is conducive to between PGC-1α and cancer metastasis needs to be fur- cell adhesion and metastasis. However, the capillary ther studied. endothelial cells are tightly connected, and therefore, Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 7 of 16

Oxygen not only changes the energy metabolism metastasis of AKR1B10­ High cancer cells but did not afect of cancer cells but also exposes cancer cells to oxida- ­AKR1B10Low cancer cells [90]. Further analysis confrmed tive stress [81]. Terefore, antioxidant programmes are that AKR1B10 can limit the side efects caused by of oxi- required in lung metastatic cancer cells. Peroxiredoxin-2 dative stress, thereby maintaining FAO in lung metastatic can promote breast cancer metastasis to the lung by reg- breast cancer [90]. Proline dehydrogenase catalyses the ulating oxidative stress, and silencing peroxiredoxin-2 catabolism of proline. In a mouse model of lung metas- can inhibit breast cancer lung metastasis by inducing oxi- tasis, the expression of proline dehydrogenase in the lung dative damage [82]. Flura-seq, a highly sensitive in situ metastatic cells was higher than primary cancer cells [91]. sequencing technology, and mouse PDX breast cancer Inhibition of proline dehydrogenase efectively reduced models were used in a study showing that compared with the formation of lung metastasis in these mice [91]. breast fat pad and brain micrometastases, lung microme- Another study found that the bioavailability of asparagine tastases induce mitochondrial electron transport com- afected the metastatic potential of breast cancer [92]. plex I, oxidative stress and antioxidation programmes Knockdown of asparagine synthetase, L-asparaginase [83]. Tis study confrmed an increase in lung metasta- treatment or direct restriction of asparagine in the diet sis-specifc oxidative stress and the upregulation of anti- can reduce lung metastasis by breast cancer cells without oxidant programmes in clinical samples [83]. In fact, at afecting primary cancer growth [92]. Increasing dietary multiple stages of cancer metastasis, cancer cells must asparagine or enhancing the expression of asparagine overcome oxidative stress while breaking away from the synthetase can promote breast cancer metastasis to the extracellular matrix, living in the circulatory system, and lung [92]. Terefore, this study also reveals that the use colonizing [79, 84, 85]. Terefore, the ability of cancer of dietary therapy can be explored from the perspective cells to adapt to oxidative stress may be acquired before of cancer metabolism to reduce the incidence of clinical they metastasize to the lung. cancer metastasis. Many metabolites or metabolic enzymes can infuence In summary, metastatic cancer cells in the lung can lung metastasis through diferent mechanisms. Pyruvate adjust their energy metabolism to aerobic oxidation and carboxylase (PC) catalyses pyruvate to produce oxaloac- adapt to the high oxidative stress environment, and they etate, which can then be used for gluconeogenesis or to can be afected by certain metabolites and metabolic replenish the TCA cycle. Lung metastatic breast cancer enzymes. However, it is still unclear how the metabolic depends on PC. PC-deprived breast cancer cells show a capacity of cells in the lung and the inherent metabolic decrease in glycolytic capacity and oxygen consumption functions of the lung afect the metabolism of lung rate and an increase in sensitivity to oxidative stress [86]. metastases. In addition, because of the tight junctions of Trough 13C tracing analysis, it was found that, com- capillary endothelial cells in the lungs, metastatic breast pared with primary breast cancer, the concentration of cancer cells tend to show increased expression of pros- mitochondrial pyruvate in lung metastatic breast cancer taglandin-endoperoxide synthase 2 (PTGS2, also known increases, thereby promoting PC-dependent replenish- as COX2) [93]. Prostaglandin, a product of PTGS2, can ment through enzymatic kinetics [87]. Surprisingly, the enhance capillary permeability. Cancer metabolism growth of primary NSCLC also depends on PC [40, 88]. enhancement of lung capillary permeability to promote Terefore, factors in the lung microenvironment may cell migration across the endothelium remains to be enable both primary and metastatic cells to acquire a PC- studied. dependent phenotype. However, the factors in the lung microenvironment that promote cancer cell acquisiton of Brain a PC-dependent phenotype remain to be explored. Pyru- Te brain is protected by the blood–brain barrier and the vate can also be directly converted into α-ketoglutarate blood-cerebrospinal fuid barrier forming a relatively iso- by alanine aminotransferase [89]. Subsequently, lated environment. Although these barriers can protect α-ketoglutarate activates P4HA through enzymatic the brain from external harmful substances, they also kinetic metabolism, promotes collagen hydroxylation, block some therapeutic drugs from entering the brain extracellular matrix remodelling, and the growth of lung [94]. Neurons are the functional units of the nervous sys- metastatic breast cancer [89]. Fatty acid metabolism tem. According to the neurotransmitters released from can also afect ability of cancers to metastasize the lung. their nerve endings, neurons can be classifed into glu- High expression of AKR1B10 in breast cancer cells show tamatergic neurons and gamma-aminobutyric neurons. reduced glycolytic activity and dependence on glucose Terefore, cancer cells that can use these neurotransmit- and enhanced fatty acid oxidation (FAO) [90]. In three- ters for processes such as energy synthesis, biosynthe- dimensional cancer spheroids formed in vitro and cancer sis, and antioxidative stress have tremendous survival metastases formed in vivo, FAO inhibitors blocked the and proliferation advantages. Brain metastasis often Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 8 of 16

occurs following lung cancer, breast cancer or melanoma. nonessential amino acid synthesis pathway activity are According to the location of metastasis, secondary brain enhanced in glioma and brain metastatic tumours. How- cancer be divided into brain parenchymal metastasis ever, the study also found that less than 50% of the acetyl- and leptomeningeal metastasis [94]. Te latter is a can- CoA pool is derived from glucose in the blood, indicating cer that has metastasized to the subarachnoid space. Te that other substances supply the energy metabolism of metabolic microenvironments of the brain parenchyma these glioma and brain metastatic cells [98]. In addition and the subarachnoid space are completely diferent. Te to glucose metabolism, lipid metabolism has an impact brain parenchyma is composed of many types of cells, on brain metastasis. Te FABP family is involved in the including neurons, astrocytes, oligodendrocytes, which uptake, transport and metabolism of fatty acids. High consume a large amount of oxygen and mainly procure expression of FABP7 is associated with poor prognosis a sufcient amount of fuel from glucose or ketone bodies and a high incidence of brain metastasis in breast can- produced in the liver to undergo oxidative metabolism cer patients [99]. Trough in vivo and in vitro experi- [95]. Tus, the brain has the characteristics of low energy ments, researchers discovered that FABP7 can promote reserves, high energy supply and high energy consump- HER2 + breast cancer cells adaptation to the brain micro- tion. Te subarachnoid space is flled with cerebrospinal environment by supporting the glycolytic phenotype fuid, which is notably hypoxic and nutrient defcient [95]. and lipid droplet storage [99]. Amino acid metabolism Diferent metabolic microenvironments may also lead to can also afect brain metastasis. Brain metastatic breast diferent metabolic characteristics of brain metastases. cancer cells can use gluconeogenesis and branched chain Currently, the metabolic adaptations necessary for amino acid oxidation to promote the survival and growth brain-specifc metastasis is mainly related to brain-spe- of breast cancer in the brain microenvironment [100]. cifc metabolites and the interaction between brain cells Acetate metabolism also promotes brain metastases. and tumour cells. Proteomics analyses have revealed Both primary brain tumours and various types of brain that breast cancer cells metastasized to the brain tend to metastatic cells can oxidize acetate to meet their energy enhance the expression of metabolic enzymes involved requirements [43]. Terefore, the glucose metabolism, in glycolysis, TCA cycle, OXPHOS, pentose phosphate lipid metabolism, amino acid metabolism and acetic acid pathway and glutathione system. Tese fndings indicate metabolism of brain metastatic tumours undergo difer- that brain metastatic breast cancer cells may, on the one ent degrees of changes, indicating that brain metastatic hand, obtain energy through the oxidative metabolism tumours seem to have extraordinary metabolic fexibility, of glucose and, on the other hand, respond to oxidative facilitating the adaptation of these cells to the complex stress by enhancing the alternative glucose metabolism and unique brain microenvironment. pathway and maintaining cell redox homeostasis, thus Te most common cell type in the brain is the neu- gaining an advantage in survival and growth in the brain ron. Neurons and cancer cells can communicate through microenvironment [96]. Comparing the gene expression neurotransmitters to promote the colonization of brain profles of brain metastases and extracerebral metastases metastases. Gamma-aminobutyrate (GABA) is a com- of melanoma, the metastatic melanomas in the brain are mon neurotransmitter in the brain. Brain tumour sam- enriched in the OXPHOS pathway [97]. By comparing ples obtained during neurosurgery for HER2 + and OXPHOS gene expression, the metabolome and meta- triple-negative breast cancer, revealed that brain metas- bolic pathway activity in intracranial and subcutaneous tasis of breast cancer has GABAergic characteristics; xenograft melanoma mice, the conclusion that OXPHOS that is, the brain metastatic cells have high expression activity is increased in brain metastatic melanoma was of GABA-related proteins, including ­GABAA receptors, also confrmed [97]. OXPHOS inhibitors can improve the GABA transporters, GABA transaminase and gluta- survival rate of intracranial melanoma xenograft mice, mate decarboxylase [101]. Further in vitro studies found resistant to MAPK inhibitors and inhibit brain metastasis that brain metastatic breast cancer cells undergo GABA in spontaneous brain metastasis of melanoma in mouse catabolism to increase NADH levels, which indicates models [97]. Similarly, a study that infused 13C-labelled metastatic breast cancer cells with proliferation advan- glucose during surgical resection of brain gliomas and tages in the brain [101]. Glutamate is also a common brain metastases and then used 13C NMR spectroscopy neurotransmitter in the brain, and one of its receptors to test related metabolites found that brain metastasis is the N-methyl-D-aspartate receptor (NMDAR). In glu- and primary brain tumour metabolism are very similar. taminergic synapses, glutamate released by presynap- Tese cells mainly used glucose for mitochondrial oxida- tic neurons is rapidly absorbed by postsynaptic neurons tion and for the production of lactic acid, alanine, glu- and astrocytes around synapses, which not only ensures tamic acid, glutamine and glycine [98]. Tese fndings the timeliness of synaptic conduction but also prevents show that glycolysis, the TCA cycle, OXPHOS and some the toxic efect of high extracellular concentrations of Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 9 of 16

glutamate on nerve cells [102]. A recent study found cancer cells [109]. In a comparison of primary ovarian that breast cancer cells can replace astrocytes and form cancer and omental metastasis, FABP4 was found to be pseudo-tripartite synapses with glutamatergic neuron upregulated in omental metastatic ovarian cancer and is cells [103]. Metastatic cells can use glutamate released located at the interface between adipocytes and cancer by neuronal cells to activate NMDAR receptors, thereby cells [109]. FABP4-defcient mice with ovarian cancer promoting brain metastasis [103]. In addition to neu- have reduced omental metastasis, confrming that adi- rons, other cells in the brain can also metabolically pocyte-derived lipids play important roles in the omen- interact with brain metastatic cells through diferent tal metastasis of ovarian cancer [109]. Another study mechanisms. Astrocytes can sequester calcium ions in found that in a coculture model consisting of adipocytes the cytoplasm of melanoma cells through gap junctions and ovarian cancer cells, the expression of the fatty acid and promote the chemotherapy resistance of melanoma receptor CD36 on the membrane of the ovarian cancer cells in brain metastases [104]. Brain metastatic tumour cells is increased, promoting the uptake of fatty acids cells can also deliver the second messenger cGAMP to by ovarian cancer cells [110]. In addition, CD36 plays a astrocytes through gap junctions, activate the STING sig- role in multiple processes of ovarian cancer metastasis nalling pathway in astrocytes, and produce the infamma- to the omentum, including adhesion, invasion, migra- tory factors IFNα and TNF. Subsequently, infammatory tion and non-anchored growth [110]. Terefore, inhi- factors activate the STAT1 and NF-κB signalling path- bition of CD36 can efectively inhibit ovarian cancer ways of brain metastatic cells through paracrine signal- metastasis. Secreted protein acidic and rich in cysteine ling to promote tumour cell growth and drug resistance (SPARC) can inhibit omental metastasis of ovarian can- [105]. Terefore, brain metastatic cells can promote the cer by inhibiting the interaction between ovarian cancer process of colonization through metabolic interactions cells and adipocytes, thereby inhibiting the phenotypic with brain cells. plasticity of omental adipocytes and metabolic repro- In summary, brain metastatic cells have a high degree gramming [111]. Among the inhibitory mechanisms, of metabolic reprogramming, and they can also interact the efect of SPARC on metabolism is mainly to inhibit with neurons and astrocytes through neurotransmitters the metabolic reprogramming of adipocytes and ovarian and gap junctions. However, the metabolic interactions cancer cells, inhibit adipocyte diferentiation, and inhibit between other types of cells and brain metastatic cells adipocytes from acquiring cancer-related phenotypes remain to be studied, and the diferences in the metabo- [111]. Compared with its level in primary ovarian cancer, lism of parenchymal brain metastasis and leptomeningeal salt-induced kinase 2 (SIK2) is highly expressed in lipid- metastasis also needs to be studied. In addition, similar rich metastases [112]. Adipocytes can activate the phos- to breast cancer cells with a tendency for lung metasta- phorylation and activation of SIK2 in ovarian cancer cells sis, the expression of PTGS2 in breast cancer cells with in a calcium-dependent manner. On the one hand, acti- a tendency for brain metastasis is also increased [106], vated SIK2 can activate the PI3K-AKT signalling pathway suggesting that lung and brain metastatic breast cancer to promote cancer cell proliferation and survival; on the may pass through tight capillary connections through other hand, it can inhibit acetyl-coenzyme A carboxylase similar tumour metabolic mechanisms. In fact, a study in combination with AMPK phosphorylation, increasing found that PTGS2 is involved in the formation of cer- the expression of carnitine palmitoyl transferase 1 and ebrospinal fuid tumour cells in brain metastatic breast activating FAO, thereby promoting omental metastasis cancer, which also indicates that PTGS2 is closely related of ovarian cancer [112]. In addition to lipid metabolism, to brain metastasis [107]. omental adipose stromal cells can induce increased nitric oxide synthesis in ovarian and endometrial cancer cells, Others thereby promoting cancer cell proliferation [113]. Nitric Approximately 80% of serous ovarian cancer metasta- oxide is catalysed from arginine via nitric oxide synthase, sizes to the omentum. Te omentum is rich in adipocytes and citrulline is also produced. Research using a model and stores a large amount of lipids, which can promote of cocultured omental adipose stromal cells and cancer omentum-specifc metastasis of ovarian cancer [108]. cells found that the cancer cells absorb arginine secreted Trough the coculture model of adipocytes and ovarian by the omental adipose stromal cells and then secrete cit- cancer cells, adipocytes were found to transport lipids rulline into the microenvironment [113]. Citrulline can to ovarian cancer cells and promote cancer cell growth increase adipogenesis in omental adipose stromal cells in vivo and in vitro [109]. Further studies revealed that [113]. Further studies found that the increased nitric the coculture model can induce lipid decomposition of oxide synthesis in cancer cells induced by omental adi- adipocytes, promote β-oxidation of cancer cells, and pose stromal cells can upregulate glycolysis, reduce mito- enhance the acquisition of rich sources of energy by chondrial respiration, and reduce oxidative stress [113]. Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 10 of 16

In addition to adipocytes, cancer-associated fbroblasts Conclusions (CAFs) can also afect omental metastasis of ovarian can- Driving efects of the metabolic microenvironment cer through cancer metabolism. With a model of cocul- on cancer metastasis tured omentum-derived CAFs and ovarian cancer cells, Increasing evidence shows that the metabolic microen- it was found that the activation of p38α MAPK in the vironment plays important roles in driving metastasis. CAFs promote the phosphorylation and activation of the On the one hand, as mentioned above, cancer cells need glycogen metabolism enzyme glucose phosphate mutase to adapt to the metabolic microenvironment of the sec- 1 under normoxic conditions in the cancer cells [114]. ondary environment to survive and proliferate (Fig. 2). Subsequently, the glycogenolysis of the cancer cells is On the other hand, cancer cells that cannot adapt to increased, which supplies glycolysis and promotes cancer the primary metabolic microenvironment may leave the cell proliferation, invasion and metastasis [114]. Further original site and colonize a microenvironment that bet- studies using in vivo models showed that the depletion of ter suited to their metabolic needs. Hypoxic lung cancer p38α in CAFs and the inhibition of glycogenolysis in can- cells leave the oxygen-rich lung microenvironment and cer cells can reduce omental metastasis [114]. Terefore, metastasize to hypoxic environments [120]. Liver cancer the propensity for omentum metastasis is related to mul- cells with vigorous mitochondrial metabolism can leave tiple cell types and multiple metabolic pathways, which is the hypoxic liver microenvironment and metastasize to of great signifcance to our understanding of the propen- the oxygen-rich lung microenvironment [121]. Terefore, sity for cancer metastasis. studying the specifc metabolic characteristics of various Lymph node metastasis is often used to predict the risk tissues and organs will help us further understand the of distant metastasis and death. It is also closely related reasons for cancer metastasis to secondary sites. In addi- to the prognosis of clinical patients and the choice of tion, designing drugs based on the metabolic characteris- treatment method. A number of studies have shown tics of various tissues and organs will enable the targeting that lymph node metastasis can be used as a foothold of metastasis from a new perspective. for distant metastasis [115, 116]. However, some studies have shown that lymph node metastasis is not the only Process of mutual adaptation between seeds and the soil way for distant metastasis [117]. Metabolic reprogram- While paying attention to the diferent metabolic micro- ming studies of lymph node metastatic tumours mainly environments of diferent metastatic organs, we need to focus on lipid metabolism. In melanoma and breast can- also pay attention to the dynamic changes of the meta- cer cells metastasized to lymph nodes, accumulated bile bolic microenvironment of the metastatic organ. On the acid activates YAP through the vitamin D receptor, which one hand, primary tumours can transform the remote leads to FAO activation of the cancer cells and facilities metabolic microenvironment of secondary organs their adaptation to the lymph node microenvironment through diferent mechanisms, which is the process of [118]. Te long-chain noncoding RNA LNMICC recruits forming premetastatic niches. On the other hand, as the nuclear factor NPM1 to the promoter of FABP5 to repro- cancer reaches the secondary organs and changes from gramme fatty acid metabolism and then promotes cer- a micrometastasis to a macrometastasis, the microenvi- vical cancer cell metastasis to lymph nodes through the ronment of the secondary organs is constantly modifed. EMT and VEGF-C-mediated lymphangiogenesis [119]. Terefore, the metabolic adaptation process of organ- Terefore, lipid metabolism-targeted drugs may inhibit specifc metastasis is not merely a process in which the lymph node metastasis, thereby delaying cancer pro- cancer cells constantly change; it is also a process in gression. In fact, the use of the FAO inhibitor etomoxir, which cancer cells and metastatic organs continuously a hypoglycemic agent, can efectively inhibit lymph node adapt, infuence and change each other. metastasis [118].

(See fgure on next page.) Fig. 2 Metabolic adaptation mechanism of metastatic organotropism. Cancers that metastasize to bone (a), liver (b), lung (c) and brain (d), omentum, and lymph nodes need to adapt to diferent metabolic microenvironments to survive and grow at a distant site. Metastasizing cancer cells need to adapt to the metabolic microenvironments of the secondary organ, which mainly includes changes in energy and nutrient sources, organ-specifc metabolites, pH, the degree of hypoxia, and the metabolic interactions between organ-specifc cells and cancer cells. HA hydroxyapatite, Ser serine, Lact lactate, FA fatty acid, BMSCs bone marrow stromal cells, ALDOB aldolase B, HSCs hepatic stellate cells; HMFs hepatic myofbroblasts, OXPHOS oxidative phosphorylation, CKB brain-type creatine kinase, FAO fatty acid oxidation, PGC-1α peroxisome proliferation-activated receptor co-activator 1 alpha; PRDX2 peroxiredoxin-2, PRODH proline dehydrogenase, ASNS asparagine synthetase, PC pyruvate carboxylase, ALT alanine aminotransferase, α-KG α-ketoglutarate, ECM extracellular matrix, GABA gamma-aminobutyrate, PPP pentose phosphate pathway, Glu glutamate, BCAA​ branched chain amino acid Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 11 of 16 Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 12 of 16

Selection or adaptation? of metastatic cells [83] to gain understanding into spa- Is the metabolic adaptation mechanism of metastatic tially specifc metabolic reprogramming during metasta- cells the result of selection or adaptation? If it is the result sis. Moreover, this technology is helpful for studying the of selection, then the driving factors of the metabolic metabolic diferences of cancer cells in diferent environ- phenotype likely originate from mutations, copy number ments within the same primary or metastatic tumour. variations and other genomic changes. Terefore, we may Metabolomics and metabolic fux analysis, the two major be able to identify the driving genes critical for metabolic metabolic high-throughput analysis methods, may signif- phenotype changes in metastatic cancer cells through icantly enhance the overall capabilities for comprehensive large-scale metastatic cancer genome resequencing. research in this feld. Te combination of transcriptom- If the metabolic adaptation mechanism is the result of ics, proteomics, metabolomics and other multi-omics adaptation, than the factors driving the acquisition of the technologies will also reveal the mechanisms of cancer metabolic phenotype are reversible mechanisms, and are metastasis and metabolic reprogramming on multiple derived from the environment, likely in relation to epi- levels. genetic mechanisms. Terefore, in the process of organ- specifc metastasis, changes in the environment may Clinical signifcance of metastatic adaptation mechanisms induce epigenetic changes, leading to changes in metabo- According to the metabolic adaptation mechanism of lism. Metabolism can also afect epigenetic modifcations cancer metastasis, diferent cancer cells that metastasize [17], thereby further promoting organ-specifc metasta- to the same organ have similar metabolic characteristics. sis. Because epigenetic changes are fast and reversible Terefore, designing therapeutic targets for these meta- compared to mutations, they may be more fexible and bolic characteristics may provide new ideas for the pre- common. vention and treatment of cancer metastasis. Since liver metastasis is usually more active in anaerobic glycolysis, Treating CTCs as a form of circulatory system metastasis we can try to use respiratory chain inhibitors to prevent When studying the metabolic adaptation mechanism of the process of lung metastasis. By contrast, we can also metastasis, we can regard CTCs as a form of circulatory try to use glycolysis inhibitors to interfere liver metastasis system metastasis and explore how cancer cells match which has an active anaerobic glycolysis mode. In addi- the metabolic microenvironment of the circulatory sys- tion to targeting the metabolic process of cancer cells, it tem. After non-transformed cells detach from the extra- is also possible to try to change the inherent metabolic cellular matrix of the primary site, they often undergo characteristics of certain organs with a high frequency increased intracellular ROS and anoikis. Cancer cells can of metastasis, making it difcult for cancer cells to adapt resist this process by enhancing the intracellular antioxi- to corresponding organs. Te bone matrix degradation dant programme [122]. Terefore, the current research within the metastatic microenvironment plays a vital with metabolic models of CTCs mainly focuses on anti- role in osteolytic metastatic cancer. Terefore, inhibiting oxidant metabolism. However, treating CTCs from the this process may hinder bone metastasis. Te local acidic perspective of metabolic adaptation, we may discover environment of liver and bone may have a certain impact more possible research directions, such as the infuence on the adaptability of metastatic cells. As a result, adjust- of the diference in arterial and venous oxygen or nutri- ing the pH of local organs through oral administration ent content in the circulatory system, and the infuence of or local addition has great potential for the treatment of metabolic interactions between various cells in the circu- metastasis. In fact, existing studies have found that bicar- latory system and cancer cells. bonate can markedly enhance the anticancer activity of transarterial chemoembolization [123]. In the termi- Application of emerging technologies nal stage of clinical cancer patients, cancer cells mostly Te application of emerging technologies can produce pass blood or lymphatic vessels to reach multiple organs more in-depth insights into the metabolic adaptation throughout the body. Targeting circulatory or lymphatic mechanism of organ-specifc metastasis. Single-cell tran- metastasis may be able to prevent or inhibit this deadly scriptome sequencing combined with PDX models can process. Hence, we can try to use hemofltration or other be used to study rare cell subpopulations during cancer devices to adjust the concentration of some metastasis- metastasis. For example, studies on the micrometastatic specifc nutrients in these vessels, thereby inhibiting the clones formed by DTCs may reveal a more refned meta- process of multiorgan invasion. static process [77]. Using in situ transcriptome sequenc- ing technology combined with a PDX model not only Abbreviations enables study into the changes in the expression of meta- CTCs: Circulating tumour cells; DTCs: Disseminated tumour cells; OXPHOS: Oxi- bolic enzymes but also the spatial location information dative phosphorylation; CP: Carbamoyl phosphate; NSCLC: Non-small cell lung Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 13 of 16

cancer; CPS: Carbamoyl phosphate synthase; UMPS: Uridine monophosphate cancer: a SEER database analysis and systematic literature review. Sci synthase; HA: Hydroxyapatite; FABP: Fatty acid-binding protein; BMSCs: Bone Rep. 2020;10(1):4444. marrow stromal cells; HSCs: Hepatic stellate cells; HMFs: Hepatic myofbro- 8. Kennecke H, Yerushalmi R, Woods R, Cheang MC, Voduc D, Speers CH, blasts; SDHB: Succinate dehydrogenase B; PDECs: Pancreatic ductal epithelial Nielsen TO, Gelmon K. Metastatic behavior of breast cancer subtypes. cells; PGC-1α: Peroxisome proliferation-activated receptor co-activator 1 J Clin Oncol. 2010;28(20):3271–7. alpha; EMT: Epithelial-mesenchymal transition; PC: Pyruvate carboxylase; FAO: 9. Wu Q, Li J, Zhu S, Wu J, Chen C, Liu Q, Wei W, Zhang Y, Sun S. Breast Fatty acid oxidation; PTGS2: Prostaglandin-endoperoxide synthase 2; GABA: cancer subtypes predict the preferential site of distant metastases: a Gamma-aminobutyrate; NMDAR: N-methyl-D aspartate receptor; SPARC:​ SEER based study. Oncotarget. 2017;8(17):27990–6. Secreted protein acidic and rich in cysteine; SIK2: Salt-induced kinase 2; and 10. Chen W, Hofmann AD, Liu H, Liu X. Organotropism: new insights into CAFs: Cancer-associated fbroblasts. molecular mechanisms of breast cancer metastasis. NPJ Precis Oncol. 2018;2(1):4. Acknowledgements 11. Urosevic J, Garcia-Albeniz X, Planet E, Real S, Cespedes MV, Guiu M, The authors would like to thank Mrs. Shuhua Zhang and Mr. Yudi Yao for their Fernandez E, Bellmunt A, Gawrzak S, Pavlovic M, et al. Colon cancer insightful comments during their review of this manuscript. cells colonize the lung from established liver metastases through p38 MAPK signalling and PTHLH. Nat Cell Biol. 2014;16(7):685–94. Authors’ contributions 12. Nguyen DX, Bos PD, Massague J. Metastasis: from dissemination to CW carried out the primary literature search and drafted the manuscript; DL organ-specifc colonization. Nat Rev Cancer. 2009;9(4):274–84. revised the manuscript. All authors read and approved the fnal manuscript. 13. Gao Y, Bado I, Wang H, Zhang W, Rosen JM, Zhang XH. Metas- tasis organotropism: redefning the congenial soil. Dev Cell. Funding 2019;49(3):375–91. This work was supported by grants from the National Natural Science Founda- 14. Schild T, Low V, Blenis J, Gomes AP. Unique metabolic adaptations tion of China, nos. 81560464 and 31960152 (DL) and Nanchang University dictate distal organ-specifc metastatic colonization. Cancer Cell. National College Students Innovation and Entrepreneurship Training Program, 2018;33(3):347–54. nos. 2020CX270 (CW). 15. DeBerardinis RJ, Chandel NS. Fundamentals of cancer metabolism. Sci Adv. 2016;2(5):e1600200. Availability of data and materials 16. Koppenol WH, Bounds PL, Dang CV. Otto Warburg’s contribu- Not applicable. tions to current concepts of cancer metabolism. Nat Rev Cancer. 2011;11(5):325–37. 17. Pavlova NN, Thompson CB. The emerging hallmarks of cancer Declarations metabolism. Cell Metab. 2016;23(1):27–47. 18. Newman JC, Verdin E. Ketone bodies as signaling metabolites. Trends Ethics approval and consent to participate Endocrinol Metab. 2014;25(1):42–52. Not applicable. 19. Huang D, Li T, Wang L, Zhang L, Yan R, Li K, Xing S, Wu G, Hu L, Jia W, et al. Hepatocellular carcinoma redirects to ketolysis for progression Consent for publication under nutrition deprivation stress. Cell Res. 2016;26(10):1112–30. Not applicable. 20. Vriens K, Christen S, Parik S, Broekaert D, Yoshinaga K, Talebi A, Dehairs J, Escalona-Noguero C, Schmieder R, Cornfeld T, et al. Evi- Competing interests dence for an alternative fatty acid desaturation pathway increasing Chao Wang and Daya Luo declare that they have no confict of interest. This cancer plasticity. Nature. 2019;566(7744):403–6. article does not contain any studies with human or animal subjects performed 21. Keshet R, Szlosarek P, Carracedo A, Erez A. Rewiring urea cycle by any of the authors. metabolism in cancer to support anabolism. Nat Rev Cancer. 2018;18(10):634–45. Author details 22. Li L, Mao Y, Zhao L, Li L, Wu J, Zhao M, Du W, Yu L, Jiang P. p53 regula- 1 School of Basic Medical Sciences, Nanchang University, Nanchang 330006, 2 tion of ammonia metabolism through urea cycle controls polyamine China. Department of Biochemistry and Molecular Biology, School of Basic biosynthesis. Nature. 2019;567(7747):253–6. Medical Sciences, Nanchang University, Nanchang 330006, China. 23. Spinelli JB, Yoon H, Ringel AE, Jeanfavre S, Clish CB, Haigis MC. Meta- bolic recycling of ammonia via glutamate dehydrogenase supports Received: 13 December 2020 Accepted: 19 April 2021 breast cancer biomass. Science. 2017;358(6365):941–6. 24. Kim J, Hu Z, Cai L, Li K, Choi E, Faubert B, Bezwada D, Rodriguez- Canales J, Villalobos P, Lin YF, et al. CPS1 maintains pyrimidine pools and DNA synthesis in KRAS/LKB1-mutant lung cancer cells. Nature. 2017;546(7656):168–72. References 25. Lee JS, Adler L, Karathia H, Carmel N, Rabinovich S, Auslander N, 1. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Keshet R, Stettner N, Silberman A, Agemy L, et al. Urea cycle dys- Cell. 2011;144(5):646–74. regulation generates clinically relevant genomic and biochemical 2. Lambert AW, Pattabiraman DR, Weinberg RA. Emerging biological signatures. Cell. 2018;174(6):1559-1570.e1522. principles of metastasis. Cell. 2017;168(4):670–91. 26. Bi J, Wu S, Zhang W, Mischel PS. Targeting cancer’s metabolic co- 3. Obenauf AC, Massague J. Surviving at a distance: organ-specifc metas- dependencies: a landscape shaped by genotype and tissue context. tasis. Trends Cancer. 2015;1(1):76–91. Biochim Biophys Acta Rev Cancer. 2018;1870(1):76–87. 4. Liu Q, Zhang H, Jiang X, Qian C, Liu Z, Luo D. Factors involved in cancer 27. Kim J, DeBerardinis RJ. Mechanisms and implications of metabolic metastasis: a better understanding to “seed and soil” hypothesis. Mol heterogeneity in cancer. Cell Metab. 2019;30(3):434–46. Cancer. 2017;16(1):176. 28. Goncalves MD, Hopkins BD, Cantley LC. Phosphatidylinositol 3-kinase, 5. Gandaglia G, Karakiewicz PI, Briganti A, Passoni NM, Schifmann J, growth disorders, and cancer. N Engl J Med. 2018;379(21):2052–62. Trudeau V, Graefen M, Montorsi F, Sun M. Impact of the site of metas- 29. Pupo E, Avanzato D, Middonti E, Bussolino F, Lanzetti L. KRAS-driven tases on survival in patients with metastatic prostate cancer. Eur Urol. metabolic rewiring reveals novel actionable targets in cancer. Front 2015;68(2):325–34. Oncol. 2019;9:848. 6. Yang J, Manson DK, Marr BP, Carvajal RD. Treatment of 30. Commisso C, Davidson SM, Soydaner-Azeloglu RG, Parker SJ, Kam- uveal melanoma: where are we now? Ther Adv Med Oncol. phorst JJ, Hackett S, Grabocka E, Nofal M, Drebin JA, Thompson CB, 2018;10:1758834018757175. et al. Macropinocytosis of protein is an amino acid supply route in 7. Liu Q, Zhang R, Michalski CW, Liu B, Liao Q, Kleef J. Surgery for syn- Ras-transformed cells. Nature. 2013;497(7451):633–7. chronous and metachronous single-organ metastasis of pancreatic Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 14 of 16

31. Stine ZE, Walton ZE, Altman BJ, Hsieh AL, Dang CV. MYC, metabolism, 51. Saidak Z, Boudot C, Abdoune R, Petit L, Brazier M, Mentaverri R, Kamel and cancer. Cancer Discov. 2015;5(10):1024–39. S. Extracellular calcium promotes the migration of breast cancer cells 32. Kruiswijk F, Labuschagne CF, Vousden KH. p53 in survival, death and through the activation of the calcium sensing receptor. Exp Cell Res. metabolic health: a lifeguard with a licence to kill. Nat Rev Mol Cell Biol. 2009;315(12):2072–80. 2015;16(7):393–405. 52. Joeckel E, Haber T, Prawitt D, Junker K, Hampel C, Thurof JW, Roos 33. Dang L, White DW, Gross S, Bennett BD, Bittinger MA, Driggers EM, FC, Brenner W. High calcium concentration in bones promotes bone Fantin VR, Jang HG, Jin S, Keenan MC, et al. Cancer-associated IDH1 metastasis in renal cell carcinomas expressing calcium-sensing recep- mutations produce 2-hydroxyglutarate. Nature. 2009;462(7274):739–44. tor. Mol Cancer. 2014;13:42. 34. Ward PS, Patel J, Wise DR, Abdel-Wahab O, Bennett BD, Coller HA, 53. Nakazawa MS, Keith B, Simon MC. Oxygen availability and metabolic Cross JR, Fantin VR, Hedvat CV, Perl AE, et al. The common feature adaptations. Nat Rev Cancer. 2016;16(10):663–73. of leukemia-associated IDH1 and IDH2 mutations is a neomorphic 54. Hiraga T, Kizaka-Kondoh S, Hirota K, Hiraoka M, Yoneda T. Hypoxia and enzyme activity converting alpha-ketoglutarate to 2-hydroxyglutarate. hypoxia-inducible factor-1 expression enhance osteolytic bone metas- Cancer Cell. 2010;17(3):225–34. tases of breast cancer. Cancer Res. 2007;67(9):4157–63. 35. Xiao M, Yang H, Xu W, Ma S, Lin H, Zhu H, Liu L, Liu Y, Yang C, Xu Y, et al. 55. Cox TR, Rumney RMH, Schoof EM, Perryman L, Hoye AM, Agrawal A, Inhibition of alpha-KG-dependent histone and DNA demethylases by Bird D, Latif NA, Forrest H, Evans HR, et al. The hypoxic cancer secretome fumarate and succinate that are accumulated in mutations of FH and induces pre-metastatic bone lesions through lysyl oxidase. Nature. SDH tumor suppressors. Genes Dev. 2012;26(12):1326–38. 2015;522(7554):106–10. 36. Hu J, Locasale JW, Bielas JH, O’Sullivan J, Sheahan K, Cantley LC, 56. Yoneda T, Hiasa M, Nagata Y, Okui T, White F. Contribution of acidic Vander Heiden MG, Vitkup D. Heterogeneity of tumor-induced gene extracellular microenvironment of cancer-colonized bone to bone pain. expression changes in the human metabolic network. Nat Biotechnol. Biochim Biophys Acta. 2015;1848:2677–84. 2013;31(6):522–9. 57. Ogawa T, Ishida-Kitagawa N, Tanaka A, Matsumoto T, Hirouchi T, Aki- 37. Gaude E, Frezza C. Tissue-specifc and convergent metabolic trans- maru M, Tanihara M, Yogo K, Takeya T. A novel role of L-serine (L-Ser) for formation of cancer correlates with metastatic potential and patient the expression of nuclear factor of activated T cells (NFAT)2 in receptor survival. Nat Commun. 2016;7:13041. activator of nuclear factor kappa B ligand (RANKL)-induced osteoclas- 38. Yuneva MO, Fan TW, Allen TD, Higashi RM, Ferraris DV, Tsukamoto togenesis in vitro. J Bone Miner Metab. 2006;24(5):373–9. T, Mates JM, Alonso FJ, Wang C, Seo Y, et al. The metabolic profle of 58. Pollari S, Kakonen SM, Edgren H, Wolf M, Kohonen P, Sara H, Guise T, tumors depends on both the responsible genetic lesion and tissue Nees M, Kallioniemi O. Enhanced serine production by bone metastatic type. Cell Metab. 2012;15(2):157–70. breast cancer cells stimulates osteoclastogenesis. Breast Cancer Res 39. Mayers JR, Torrence ME, Danai LV, Papagiannakopoulos T, Davidson Treat. 2011;125(2):421–30. SM, Bauer MR, Lau AN, Ji BW, Dixit PD, Hosios AM, et al. Tissue of origin 59. Lemma S, Di Pompo G, Porporato PE, Sboarina M, Russell S, Gillies RJ, dictates branched-chain amino acid metabolism in mutant Kras-driven Baldini N, Sonveaux P, Avnet S. MDA-MB-231 breast cancer cells fuel cancers. Science. 2016;353(6304):1161–5. osteoclast metabolism and activity: a new rationale for the pathogen- 40. Davidson SM, Papagiannakopoulos T, Olenchock BA, Heyman JE, Keibler esis of osteolytic bone metastases. Biochim Biophys Acta Mol Basis Dis. MA, Luengo A, Bauer MR, Jha AK, O’Brien JP, Pierce KA, et al. Environ- 2017;1863(12):3254–64. ment impacts the metabolic dependencies of ras-driven non-small cell 60. Wang H, Tian L, Liu J, Goldstein A, Bado I, Zhang W, Arenkiel BR, Li Z, lung cancer. Cell Metab. 2016;23(3):517–28. Yang M, Du S, et al. The osteogenic niche is a calcium reservoir of bone 41. Cantor JR, Abu-Remaileh M, Kanarek N, Freinkman E, Gao X, Louissaint A micrometastases and confers unexpected therapeutic vulnerability. Jr, Lewis CA, Sabatini DM. Physiologic medium rewires cellular metabo- Cancer Cell. 2018;34(5):823-839.e827. lism and reveals uric acid as an endogenous inhibitor of UMP synthase. 61. Gazi E, Gardner P, Lockyer NP, Hart CA, Brown MD, Clarke NW. Direct Cell. 2017;169(2):258-272.e217. evidence of lipid translocation between adipocytes and prostate 42. Hensley CT, Faubert B, Yuan Q, Lev-Cohain N, Jin E, Kim J, Jiang L, Ko cancer cells with imaging FTIR microspectroscopy. J Lipid Res. B, Skelton R, Loudat L, et al. Metabolic Heterogeneity in Human Lung 2007;48(8):1846–56. Tumors. Cell. 2016;164(4):681–94. 62. Brown MD, Hart C, Gazi E, Gardner P, Lockyer N, Clarke N. Infuence of 43. Mashimo T, Pichumani K, Vemireddy V, Hatanpaa KJ, Singh DK, Sirasana- omega-6 PUFA arachidonic acid and bone marrow adipocytes on met- gandla S, Nannepaga S, Piccirillo SG, Kovacs Z, Foong C, et al. Acetate is astatic spread from prostate cancer. Br J Cancer. 2010;102(2):403–13. a bioenergetic substrate for human glioblastoma and brain metastases. 63. Herroon MK, Rajagurubandara E, Hardaway AL, Powell K, Turchick A, Cell. 2014;159(7):1603–14. Feldmann D, Podgorski I. Bone marrow adipocytes promote tumor 44. Spencer JA, Ferraro F, Roussakis E, Klein A, Wu J, Runnels JM, Zaher growth in bone via FABP4-dependent mechanisms. Oncotarget. W, Mortensen LJ, Alt C, Turcotte R, et al. Direct measurement of local 2013;4(11):2108–23. oxygen concentration in the bone marrow of live animals. Nature. 64. Diedrich JD, Rajagurubandara E, Herroon MK, Mahapatra G, Hutte- 2014;508(7495):269–73. mann M, Podgorski I. Bone marrow adipocytes promote the Warburg 45. Zhang W, Bado I, Wang H, Lo HC, Zhang XH. Bone metastasis: fnd your phenotype in metastatic prostate tumors via HIF-1alpha activation. niche and ft in. Trends Cancer. 2019;5(2):95–110. Oncotarget. 2016;7(40):64854–77. 46. Fornetti J, Welm AL, Stewart SA. Understanding the Bone In Cancer 65. Dai J, Escara-Wilke J, Keller JM, Jung Y, Taichman RS, Pienta KJ, Keller Metastasis. J Bone Miner Res. 2018;33(12):2099–113. ET. Primary prostate cancer educates bone stroma through exoso- 47. He F, Chiou AE, Loh HC, Lynch M, Seo BR, Song YH, Lee MJ, Hoerth R, mal pyruvate kinase M2 to promote bone metastasis. J Exp Med. Bortel EL, Willie BM, et al. Multiscale characterization of the mineral 2019;216(12):2883–99. phase at skeletal sites of breast cancer metastasis. Proc Natl Acad Sci U 66. Augustin HG, Koh GY. Organotypic vasculature: From descriptive het- S A. 2017;114(40):10542–7. erogeneity to functional pathophysiology. Science. 2017. https://​doi.​ 48. O’Grady S, Morgan MP. Microcalcifcations in breast cancer: from org/​10.​1126/​scien​ce.​aal23​79. pathophysiology to diagnosis and prognosis. Biochim Biophys Acta Rev 67. Brodt P. Role of the microenvironment in liver metastasis: from pre- to Cancer. 2018;1869(2):310–20. prometastatic niches. Clin Cancer Res. 2016;22(24):5971–82. 49. Cooke MM, McCarthy GM, Sallis JD, Morgan MP. Phosphocitrate inhibits 68. Rossetto A, De Re V, Stefan A, Ravaioli M, Miolo G, Leone P, Racanelli V, calcium hydroxyapatite induced mitogenesis and upregulation of Uzzau A, Baccarani U, Cescon M. Carcinogenesis and metastasis in liver. matrix metalloproteinase-1, interleukin-1beta and cyclooxygenase-2 Cancers (Basel). 2019;11(11):1731. mRNA in human breast cancer cell lines. Breast Cancer Res Treat. 69. Jang C, Hui S, Lu W, Cowan AJ, Morscher RJ, Lee G, Liu W, Tesz GJ, Birn- 2003;79(2):253–63. baum MJ, Rabinowitz JD. The small intestine converts dietary fructose 50. Liao J, Schneider A, Datta NS, McCauley LK. Extracellular calcium as a into glucose and organic acids. Cell Metab. 2018;27(2):351-361.e353. candidate mediator of prostate cancer skeletal metastasis. Cancer Res. 70. Bu P, Chen KY, Xiang K, Johnson C, Crown SB, Rakhilin N, Ai Y, Wang L, 2006;66(18):9065–73. Xi R, Astapova I, et al. Aldolase B-mediated fructose metabolism drives Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 15 of 16

metabolic reprogramming of colon cancer liver metastasis. Cell Metab. 90. van Weverwijk A, Koundouros N, Iravani M, Ashenden M, Gao Q, Poulo- 2018;27(6):1249-1262.e1244. giannis G, Jungwirth U, Isacke CM. Metabolic adaptability in metastatic 71. Teng S, Li YE, Yang M, Qi R, Huang Y, Wang Q, Zhang Y, Chen S, Li S, Lin breast cancer by AKR1B10-dependent balancing of glycolysis and fatty K, et al. Tissue-specifc transcription reprogramming promotes liver acid oxidation. Nat Commun. 2019;10(1):2698. metastasis of colorectal cancer. Cell Res. 2020;30(1):34–49. 91. Elia I, Broekaert D, Christen S, Boon R, Radaelli E, Orth MF, Verfaillie C, 72. Loo JM, Scherl A, Nguyen A, Man FY, Weinberg E, Zeng Z, Saltz L, Paty Grunewald TGP, Fendt SM. Proline metabolism supports metastasis PB, Tavazoie SF. Extracellular metabolic energetics can promote cancer formation and could be inhibited to selectively target metastasizing progression. Cell. 2015;160(3):393–406. cancer cells. Nat Commun. 2017;8:15267. 73. Dupuy F, Tabaries S, Andrzejewski S, Dong Z, Blagih J, Annis MG, 92. Knott SRV, Wagenblast E, Khan S, Kim SY, Soto M, Wagner M, Turgeon Omeroglu A, Gao D, Leung S, Amir E, et al. PDK1-dependent metabolic MO, Fish L, Erard N, Gable AL, et al. Asparagine bioavailability governs reprogramming dictates metastatic potential in breast cancer. Cell metastasis in a model of breast cancer. Nature. 2018;554(7692):378–81. Metab. 2015;22(4):577–89. 93. Minn AJ, Gupta GP, Siegel PM, Bos PD, Shu W, Giri DD, Viale A, Olshen 74. Fabian A, Stegner S, Miarka L, Zimmermann J, Lenk L, Rahn S, Buttlar AB, Gerald WL, Massague J. Genes that mediate breast cancer metasta- J, Viol F, Knaack H, Esser D, et al. Metastasis of pancreatic cancer: An sis to lung. Nature. 2005;436(7050):518–24. uninfamed liver micromilieu controls cell growth and cancer stem cell 94. Achrol AS, Rennert RC, Anders C, Sofetti R, Ahluwalia MS, Nayak L, properties by oxidative phosphorylation in pancreatic ductal epithelial Peters S, Arvold ND, Harsh GR, Steeg PS, et al. Brain metastases. Nat Rev cells. Cancer Lett. 2019;453:95–106. Dis Primers. 2019;5(1):5. 75. Altorki NK, Markowitz GJ, Gao D, Port JL, Saxena A, Stiles B, McGraw T, 95. Boire A, Brastianos PK, Garzia L, Valiente M. Brain metastasis. Nat Rev Mittal V. The lung microenvironment: an important regulator of tumour Cancer. 2020;20(1):4–11. growth and metastasis. Nat Rev Cancer. 2019;19(1):9–31. 96. Chen EI, Hewel J, Krueger JS, Tiraby C, Weber MR, Kralli A, Becker K, Yates 76. Alvarado A, Arce I. Metabolic functions of the lung, disorders and asso- JR 3rd, Felding-Habermann B. Adaptation of energy metabolism in ciated pathologies. J Clin Med Res. 2016;8(10):689–700. breast cancer brain metastases. Cancer Res. 2007;67(4):1472–86. 77. Davis RT, Blake K, Ma D, Gabra MBI, Hernandez GA, Phung AT, Yang 97. Fischer GM, Jalali A, Kircher DA, Lee WC, McQuade JL, Haydu LE, Joon Y, Maurer D, Lefebvre A, Alshetaiwi H, et al. Transcriptional diversity AY, Reuben A, de Macedo MP, Carapeto FCL, et al. Molecular profling and bioenergetic shift in human breast cancer metastasis revealed by reveals unique immune and metabolic features of melanoma brain single-cell RNA sequencing. Nat Cell Biol. 2020;22(3):310–20. metastases. Cancer Discov. 2019;9(5):628–45. 78. Bost F, Kaminski L. The metabolic modulator PGC-1alpha in cancer. Am 98. Maher EA, Marin-Valencia I, Bachoo RM, Mashimo T, Raisanen J, J Cancer Res. 2019;9(2):198–211. Hatanpaa KJ, Jindal A, Jefrey FM, Choi C, Madden C, et al. Metabo- 79. LeBleu VS, O’Connell JT, Gonzalez Herrera KN, Wikman H, Pantel K, lism of [U-13 C]glucose in human brain tumors in vivo. NMR Biomed. Haigis MC, de Carvalho FM, Damascena A, Domingos Chinen LT, Rocha 2012;25(11):1234–44. RM, et al. PGC-1alpha mediates mitochondrial biogenesis and oxidative 99. Cordero A, Kanojia D, Miska J, Panek WK, Xiao A, Han Y, Bonamici N, phosphorylation in cancer cells to promote metastasis. Nat Cell Biol. Zhou W, Xiao T, Wu M, et al. FABP7 is a key metabolic regulator in 2014. https://​doi.​org/​10.​1038/​ncb30​39. HER2 breast cancer brain metastasis. Oncogene. 2019;38(37):6445–60. 80. Andrzejewski S, Klimcakova E, Johnson RM, Tabaries S, Annis MG, 100. Chen+ J, Lee HJ, Wu X, Huo L, Kim SJ, Xu L, Wang Y, He J, Bollu LR, Gao G, McGuirk S, Northey JJ, Chenard V, Sriram U, Papadopoli DJ, et al. PGC- et al. Gain of glucose-independent growth upon metastasis of breast 1alpha promotes breast cancer metastasis and confers bioenergetic cancer cells to the brain. Cancer Res. 2015;75(3):554–65. fexibility against metabolic drugs. Cell Metab. 2017;26(5):778-787.e775. 101. Neman J, Termini J, Wilczynski S, Vaidehi N, Choy C, Kowolik CM, Li H, 81. Halliwell B. Cell culture, oxidative stress, and antioxidants: avoiding Hambrecht AC, Roberts E, Jandial R. Human breast cancer metastases pitfalls. Biomed J. 2014;37(3):99–105. to the brain display GABAergic properties in the neural niche. Proc Natl 82. Stresing V, Baltziskueta E, Rubio N, Blanco J, Arriba MC, Valls J, Janier M, Acad Sci U S A. 2014;111(3):984–9. Clezardin P, Sanz-Pamplona R, Nieva C, et al. Peroxiredoxin 2 specif- 102. Danbolt NC. Glutamate uptake. Prog Neurobiol. 2001;65(1):1–105. cally regulates the oxidative and metabolic stress response of human 103. Zeng Q, Michael IP, Zhang P, Saghafnia S, Knott G, Jiao W, McCabe metastatic breast cancer cells in lungs. Oncogene. 2013;32(6):724–35. BD, Galvan JA, Robinson HPC, Zlobec I, et al. Synaptic proximity 83. Basnet H, Tian L, Ganesh K, Huang YH, Macalinao DG, Brogi E, Finley LW, enables NMDAR signalling to promote brain metastasis. Nature. Massague J. Flura-seq identifes organ-specifc metabolic adaptations 2019;573(7775):526–31. during early metastatic colonization. Elife. 2019. https://​doi.​org/​10.​ 104. Lin Q, Balasubramanian K, Fan D, Kim SJ, Guo L, Wang H, Bar-Eli M, 7554/​eLife.​43627. Aldape KD, Fidler IJ. Reactive astrocytes protect melanoma cells from 84. Schafer ZT, Grassian AR, Song L, Jiang Z, Gerhart-Hines Z, Irie HY, chemotherapy by sequestering intracellular calcium through gap junc- Gao S, Puigserver P, Brugge JS. Antioxidant and oncogene rescue tion communication channels. Neoplasia. 2010;12(9):748–54. of metabolic defects caused by loss of matrix attachment. Nature. 105. Chen Q, Boire A, Jin X, Valiente M, Er EE, Lopez-Soto A, Jacob L, Patwa R, 2009;461(7260):109–13. Shah H, Xu K, et al. Carcinoma-astrocyte gap junctions promote brain 85. Piskounova E, Agathocleous M, Murphy MM, Hu Z, Huddlestun SE, metastasis by cGAMP transfer. Nature. 2016;533(7604):493–8. Zhao Z, Leitch AM, Johnson TM, DeBerardinis RJ, Morrison SJ. Oxidative 106. Bos PD, Zhang XH, Nadal C, Shu W, Gomis RR, Nguyen DX, Minn AJ, van stress inhibits distant metastasis by human melanoma cells. Nature. de Vijver MJ, Gerald WL, Foekens JA, et al. Genes that mediate breast 2015;527(7577):186–91. cancer metastasis to the brain. Nature. 2009;459(7249):1005–9. 86. Shinde A, Wilmanski T, Chen H, Teegarden D, Wendt MK. Pyruvate car- 107. Allen JE, Patel AS, Prabhu VV, Dicker DT, Sheehan JM, Glantz MJ, boxylase supports the pulmonary tropism of metastatic breast cancer. El-Deiry WS. COX-2 drives metastatic breast cells from brain lesions Breast Cancer Res. 2018;20(1):76. into the cerebrospinal fuid and systemic circulation. Cancer Res. 87. Christen S, Lorendeau D, Schmieder R, Broekaert D, Metzger K, Veys K, 2014;74(9):2385–90. Elia I, Buescher JM, Orth MF, Davidson SM, et al. Breast cancer-derived 108. Chkourko Gusky H, Diedrich J, MacDougald OA, Podgorski I. Omen- lung metastases show increased pyruvate carboxylase-dependent tum and bone marrow: how adipocyte-rich organs create tumour anaplerosis. Cell Rep. 2016;17(3):837–48. microenvironments conducive for metastatic progression. Obes Rev. 88. Sellers K, Fox MP, Bousamra M 2nd, Slone SP, Higashi RM, Miller DM, 2016;17(11):1015–29. Wang Y, Yan J, Yuneva MO, Deshpande R, et al. Pyruvate carboxylase 109. Nieman KM, Kenny HA, Penicka CV, Ladanyi A, Buell-Gutbrod R, Zillhardt is critical for non-small-cell lung cancer proliferation. J Clin Invest. MR, Romero IL, Carey MS, Mills GB, Hotamisligil GS, et al. Adipocytes 2015;125(2):687–98. promote ovarian cancer metastasis and provide energy for rapid tumor 89. Elia I, Rossi M, Stegen S, Broekaert D, Doglioni G, van Gorsel M, Boon R, growth. Nat Med. 2011;17(11):1498–503. Escalona-Noguero C, Torrekens S, Verfaillie C, et al. Breast cancer cells 110. Ladanyi A, Mukherjee A, Kenny HA, Johnson A, Mitra AK, Sundaresan S, rely on environmental pyruvate to shape the metastatic niche. Nature. Nieman KM, Pascual G, Benitah SA, Montag A, et al. Adipocyte-induced 2019;568(7750):117–21. CD36 expression drives ovarian cancer progression and metastasis. Oncogene. 2018;37(17):2285–301. Wang and Luo Exp Hematol Oncol (2021) 10:30 Page 16 of 16

111. John B, Naczki C, Patel C, Ghoneum A, Qasem S, Salih Z, Said N. Regula- 118. Lee CK, Jeong SH, Jang C, Bae H, Kim YH, Park I, Kim SK, Koh GY. Tumor tion of the bi-directional cross-talk between ovarian cancer cells and metastasis to lymph nodes requires YAP-dependent metabolic adapta- adipocytes by SPARC. Oncogene. 2019;38(22):4366–83. tion. Science. 2019;363(6427):644–9. 112. Miranda F, Mannion D, Liu S, Zheng Y, Mangala LS, Redondo C, Herrero- 119. Shang C, Wang W, Liao Y, Chen Y, Liu T, Du Q, Huang J, Liang Y, Liu J, Gonzalez S, Xu R, Taylor C, Chedom DF, et al. Salt-inducible kinase 2 Zhao Y, et al. LNMICC promotes nodal metastasis of cervical cancer by couples ovarian cancer cell metabolism with survival at the adipocyte- reprogramming fatty acid metabolism. Cancer Res. 2018;78(4):877–90. rich metastatic niche. Cancer Cell. 2016;30(2):273–89. 120. Hung PF, Hong TM, Chang CC, Hung CL, Hsu YL, Chang YL, Wu CT, 113. Salimian Rizi B, Caneba C, Nowicka A, Nabiyar AW, Liu X, Chen K, Klopp Chang GC, Chan NL, Yu SL, et al. Hypoxia-induced Slug SUMOylation A, Nagrath D. Nitric oxide mediates metabolic coupling of omentum- enhances lung cancer metastasis. J Exp Clin Cancer Res. 2019;38(1):5. derived adipose stroma to ovarian and endometrial cancer cells. Cancer 121. Zhang Z, Li TE, Chen M, Xu D, Zhu Y, Hu BY, Lin ZF, Pan JJ, Wang X, Wu Res. 2015;75(2):456–71. C, et al. MFN1-dependent alteration of mitochondrial dynamics drives 114. Curtis M, Kenny HA, Ashcroft B, Mukherjee A, Johnson A, Zhang Y, hepatocellular carcinoma metastasis by glucose metabolic reprogram- Helou Y, Batlle R, Liu X, Gutierrez N, et al. Fibroblasts mobilize tumor ming. Br J Cancer. 2020;122(2):209–20. cell glycogen to promote proliferation and metastasis. Cell Metab. 122. Elia I, Doglioni G, Fendt SM. Metabolic hallmarks of metastasis forma- 2019;29(1):141-155.e149. tion. Trends Cell Biol. 2018;28(8):673–84. 115. Pereira ER, Kedrin D, Seano G, Gautier O, Meijer EFJ, Jones D, Chin SM, 123. Chao M, Wu H, Jin K, Li B, Wu J, Zhang G, Yang G, Hu X. A nonran- Kitahara S, Bouta EM, Chang J, et al. Lymph node metastases can invade domized cohort and a randomized study of local control of large local blood vessels, exit the node, and colonize distant organs in mice. hepatocarcinoma by targeting intratumoral lactic acidosis. Elife. 2016. Science. 2018;359(6382):1403–7. https://​doi.​org/​10.​7554/​eLife.​15691. 116. Brown M, Assen FP, Leithner A, Abe J, Schachner H, Asfour G, Bago- Horvath Z, Stein JV, Uhrin P, Sixt M, et al. Lymph node blood vessels provide exit routes for metastatic tumor cell dissemination in mice. Publisher’s Note Science. 2018;359(6382):1408–11. Springer Nature remains neutral with regard to jurisdictional claims in pub- 117. Naxerova K, Reiter JG, Brachtel E, Lennerz JK, van de Wetering M, lished maps and institutional afliations. Rowan A, Cai T, Clevers H, Swanton C, Nowak MA, et al. Origins of lymphatic and distant metastases in human colorectal cancer. Science. 2017;357(6346):55–60.

Ready to submit your research ? Choose BMC and benefit from:

• fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year

At BMC, research is always in progress.

Learn more biomedcentral.com/submissions