Proliferating cells, such as cancer cells and activated lymphocytes, are metabolically different from nonproliferating cells as they are programmed to utilize either anaerobic or aerobic glycolysis, the Warburg effect, depending on the availability of oxygen [1]. Glucose and fructose are hexose sugars that provide glycolytic intermediates for further metabolism via the pentose cycle, tricarboxylic acid cycle (TCA), one-carbon metabolism and hexosamine biosynthesis pathway. Glycolysis is a physiological response of tissues to hypoxia, a low oxygen environment, with cells of tumors taking up glucose and producing intermediate products of glycolysis (including fructose-6-phosphate, F6P) and significant amounts of lactate. By switching from oxidative phosphorylation to glycolysis, cells rapidly generate ATP as compared with oxidation of glucose and activation of the TCA cycle. Further, glycolytic intermediates are substrates for other metabolic pathways required to meet demands for proliferation, migration, and differentiation of cells [2–5].

This review focuses on the metabolism of glucose and fructose, two molecules present in great abundances in conceptuses (embryo and extra-embryonic membranes; fetus and placenta) of ungulates and cetaceans throughout pregnancy [see [6]; Figure 1]. Fructose and lactate have been considered metabolic wastes, so their functional roles in conceptus development have not been established. The following sections of this review evidence for important roles of fructose and lactate in conceptus development. Thus, the focus of this review is on contributions of glucose and fructose to the major metabolic pathways required for development of conceptuses under oxygenated and low oxygen conditions, expression of enzymes for production and metabolism of fructose from glucose, and characterization of lactate synthesis and transport throughout pregnancy.

Another focus of this review is the role of fructose in uncontrolled growth of cancers, as Burton et al. [7] reported similarities between metabolism in placentae and in malignant tumors while acknowledging that growth of the placenta is regulated. They note that the availability of oxygen controls placental development and tumor behavior as both develop in a low oxygen environment, and both need to stimulate vascular development to deliver nutrients and oxygen sufficient to support high rates of cell proliferation. Thus, glycolysis is common in support of growth and development of both placentae and tumors. Glycolysis, rather than oxidative phosphorylation, has the advantage of maintaining carbon skeletons for the synthesis of nucleotides, cell membranes and organelles, as well as protecting the conceptus from adverse effects of free radicals.

The link between cancer and altered glucose metabolism was discovered by Otto Warburg over 100 years ago. A connection between cancer and fructose metabolism was also discovered as fructose, like glucose, affects growth, proliferation, and survival of cancer cells [8, 9]. Cancer cells express solute carrier family member 5 (SLC2A5), the fructose transporter, and upregulation of SLC2A5 is indicative of a poor prognosis for cancer patients that experienced an increase in cancer cell proliferation, colony growth, and metastasis [10]. There is over-expression of SLC2A5 for glioblastoma, colon, liver, lung, breast, and prostate cancers [11, 12].

The polyol pathway involves two steps for conversion of glucose to fructose [13–15]. Glucose is reduced to sorbitol by aldose reductase (AKR1B1) that requires nicotinamide adenine dinucleotide phosphate hydrogen (NADPH), and sorbitol is oxidized to fructose by sorbitol dehydrogenase (SORD) yielding nicotinamide adenine dinucleotide hydrogen (NADH) from nicotinamide adenine dinucleotide (NAD). The conversion of glucose to fructose may enhance glycolysis with conversion of fructose to fructose-1-PO₄ (F1P) by ketohexokinase (KHK) occurring mainly in the liver [16, 17]. KHK is also frequently expressed in tumors [18–20] that also produce fructose 1-P endogenously from fructose (a product of glucose). This reaction bypasses the rate-limiting step in glycolysis at the level of phosphofructokinase (PFK), which is inhibited by high levels of ATP, citrate and low pH. Naked mole rats also increase the production of fructose and F1P to increase their survival under low oxygen levels and even anoxia because fructose supports glycolysis and generation of ATP and substrates for other major metabolic pathways [21]. Low concentrations of oxygen inhibit oxidative phosphorylation that lowers pH and inhibits PFK. Thus, metabolism of fructose to F1P by KHK bypasses the PFK feedback inhibition by low pH, whereas glaceraldehyde-3-PO₄ (GAP) and dihydroxyacetone-PO₄ (DHAP) enter glycolysis downstream of PFK.

Most tumors develop initially in a hypoxic and acidic environment [22]. A direct link between polyol pathway activity and cancer was reported by Schwab et al. [23] who found a strong correlation between expression of AKR1B1and epithelial-to-mesenchymal transition (EMT) in patients with lung cancer and in an EMT-mediated colon cancer mouse model. They also found that AKR1B1 knockdown decreased EMT in cancers. Knockdown of expression of SORD also suppressed EMT in the cancer cell line. Schwab and colleagues [23] reported that glucose metabolism via the polyol pathway controls EMT via transforming growth factor beta (TGFB) autocrine stimulation and that expression of TGFB decreased following knockdown of AKR1B1 and SORD while EMT markers were rescued by TGFB [23]. There is also a link between the polyol pathway, particularly AKR1B1 overexpression, and development of breast, ovarian, cervical, and rectal cancers [24]. For colorectal cancer cells, AKR1B1 affects cellular proliferation and cell cycle progression, cell motility and expression of nuclear factor kappa-light-chain-enhancer of activated B cells (NFKB) resulting in a poor prognosis for colorectal cancer patients [25].

There is considerable evidence for endogenous production of fructose via the polyol pathway in cancer cells and phosphorylation of fructose to F1P by KHK in tumors of humans. This pathway for glycolysis is clearly advantageous as it bypasses PFK regulation to allow endogenously produced fructose to be used to produce F1P and substrates from the glycolytic pathway to support of cancer cells in a low oxygen, low pH environment that include the hexosamine biosynthesis pathway [26], the pentose cycle (PC) [27], de novo lipogenesis [28], one-carbon metabolism [29], and the TCA cycle [29].

The polyol pathway involves two steps for conversion of glucose to fructose [13–15] as reviewed in Fructose Metabolism and Cancer. But the first report of fructose in tissues of conceptuses was in 1855 by Bernard [30]. Between that time and 1956, numerous reports confirmed the presence or absence of fructose in blood, amniotic fluid and allantoic fluid of ungulates and other species (see Goodwin [31]). Due to confusion about the presence or absence of fructose in fetal blood among species, Goodwin [31] measured fructose in blood from ungulates (ox, goat, horse, pig), cetaceans (fin, humpback and blue whales), and nonungulates (guinea pig, rabbit, rat, dog, cat, ferret) and cited work indicating the absence of fructose in blood from newborn human babies. Thus, he divided common mammals into two groups: those with detectable fructose (ungulates and cetaceans) and those with very low or undetectable fructose in fetal blood (nonungulates and humans). It was also noted that samples of fetal blood and amniotic fluids from ungulates and cetaceans had much greater concentrations of fructose than glucose.

There are species with invasive implantation and either hemochorial (humans, monkeys and rodents) or endotheliochorial (carnivores) placentae with close apposition between maternal and blood and fetal blood for efficient transport of glucose. Species with those types of placentae have little or no fructose in fetal blood or fetal fluids. However, species with noninvasive implantation have either epitheliochorial placentae (pigs, horses, cetaceans) or syndesmochorial placentae (cows, sheep, goats) with five or six layers of cells between maternal and fetal blood. Those species have fructogenic placenta with fructose being the most abundant hexose sugar in blood and fetal fluids of those species. Fructose is present in fluid of the extra-embryonic coelom of human conceptuses in early gestation [32], but it is a minor sugar compared with glucose. Fructose is also a minor sugar in fetal blood and fetal fluids of dogs, cats, guinea pigs, rabbits, rats, and ferrets [31].

Studies of pregnant ewes [see [6, 33, 34]] revealed that: 1) injection of glucose into ewes to make them hyperglycemic resulted in a rapid increase in fructose in fetal blood; 2) injection of glucose into the fetal vasculature increased glucose in maternal blood and the fetus experienced hyperfructosemia indicating that glucose, but not fructose, could be transported from the fetal-placental vasculature to maternal blood; 3) the placenta is responsible for converting glucose to fructose; and 4) the flux of glucose from maternal blood to fetal blood may reach 70 mg/min in hyperglycemic ewes. Radiolabeled glucose was also shown to be converted to radiolabeled fructose by placentae of pigs. One may speculate about four main points. First, in species with hemochorial and endotheliochorial placenta, the chorion is in very close proximity to maternal blood for highly efficient exchange of glucose and other nutrients, as well as gases. The number of tissue layers separating maternal blood and chorion is three in species with hemochorial placenta, four in species with endotheliochorial placenta, but six in species with syndesmochorial and epitheliochorial placenta. Second, species with hemochorial placenta do not have an allantois as it regresses, like the yolk sac, early in gestation. For carnivores with endotheliochorial placenta, there is an allantoic sac, but it is not well developed. However, for ungulates, a well-developed allantois is characteristic of the placenta. The allantois forms a sac that is a reservoir of water and nutrients (e.g., fructose and glutamine) in storage and available to be recycled into the fetal-placental circulation. Third, the rate of uterine blood flow to support a human pregnancy is estimated to be 0.5 L per minute [35] compared to 2.5 L per minute for pigs [36], 5.95 L per minute for cows [37] and 2 L per minute for sheep during late gestation. The high rates of uterine blood flow for pregnant ungulates are critical for delivery and exchange of sufficient nutrients and gases to support fetal growth and development. Fourth, the placentae of ungulates convert glucose to fructose that is sequestered within the pregnant uterus and unavailable to maternal tissues. Rather, it is metabolized via glycolysis after being phosphorylated at carbon number 1 to F1P. This fructolysis pathway is not inhibited by low pH, citrate or ATP.

Fructose in conceptuses of sheep and pigs has been ignored by fetal physiologist using those animal models in biomedical research although fructose is 15- to 30-fold greater than glucose in allantoic fluid of sheep (see Figure 2), as well as blood of ovine fetuses (data not shown). A review by Battaglia and Meschia [38] on fetal and placental metabolism does not mention fructose. Fructose was considered a waste product since it was not metabolized via the classical glycolytic pathway [38–42] even though it is 11- to 33-times more abundant than glucose in blood and allantoic fluid of fetal lambs [33]. The literature indicates the following: 1) fructose can be used for synthesis of nucleic acids and generation of reducing equivalents in the form of NADPH in fetal pigs [43] and in HeLa cells [44]; 2) neither fructose nor glucose is metabolized via the PC by the ovine placenta [45]; 3) fructose and glucose are equivalent substrates for the synthesis of neutral lipids and phospholipids in heart, liver, kidney, brain, and adipose tissue of fetal lambs [46]; 4) activities of glucose-6-phosphate (G6P) dehydrogenase, malic enzyme, and acetyl-CoA carboxylase in liver are stimulated by glucose to increase lipogenesis [47]; and 5) fructose enters adipocytes by both insulin-independent and insulin-insensitive mechanisms [48]. These lines of evidence indicated that both glucose and fructose are metabolic substrates utilized by mammalian conceptuses.

Fructose is 15- to 30-fold greater in abundance than glucose in fetal fluids and plasma throughout pregnancy in sheep, for example, highlighting the facts that ungulate placentae are fructogenic and that fructose, but not glucose, is sequestered within fetal blood and fetal fluids [[49–51]; Figure 2]. The following are key findings from research on metabolism of fructose in ungulates. First, cell-specific, and temporal changes in expression of enzymes involved in the metabolism of glucose and fructose in ovine and porcine conceptuses throughout pregnancy are well characterized [6, 34, 52–59]. Second, cell-specific and temporal expression of KHK isoforms for conversion of fructose to F1P provide a substrate for metabolism via the fructolysis pathway that bypasses the regulatory step at PFK in glycolysis for unimpeded production of key substrates for the PC (G6P) [54], hexosamine biosynthesis (F6P) [6, 59], one-carbon metabolism and serine synthesis (3-phosphoglycerate) [53], and TCA cycle (pyruvate) [54]. Third, the glycolytic pathway generates abundant lactate that is not likely used for gluconeogenesis by ovine conceptuses as they do not express the required enzymes for gluconeogenesis; glucose-6-phosphatase (G6Pase) or phosphoenolpyruvate carboxykinase (PCK) [54]. Fourth, ovine conceptuses do express aldolase, an enzyme that converts glyceraldehyde-3-phosphate (GAP) and dihydroxyacetone phosphate to F-1,6-P. The latter is dephosphorylated to F6P, which is converted to G6P via phosphoglucoisomerase (PGI) for metabolism via the PC and glycogen synthesis [54]. Fifth, there is expression of O-linked N-acetylglucosaminyltransferase (OGT) mRNA in placentomes of sheep throughout gestation and OGT glycosylates and activates proteins such as AKT (protein kinase B) to inhibit silencing of TSC2 (tuberous sclerosis complex 2) and activate mechanistic target of rapamycin (mTOR) that stimulates expression of mRNAs and proteins involved in cell proliferation and migration [59]. Sixth, F6P and glutamine are required for synthesis of UDP-GlcNAc (UDP-N-acetyl-D-glucosamine) that uses OGT to activate the mTOR pathway in ovine [59] and porcine [6] conceptuses by transferring a carbohydrate moiety to a serine or a threonine residue [59]. Like fructose, glutamine is also unusually abundant in conceptuses of ungulates. For example, concentrations of glutamine in ovine allantoic fluid are approximately 25 mM at Day 60 of gestation [60].

The G-protein-coupled receptors for lactate, GPR81 (hydroxycarboxylic receptor 1), GPR109A (hydroxycarboxylic receptor 2) and GPR109B (hydroxycarboxylic receptor 3) share high sequence homology and are designated HCAR1, HCAR2 and HCAR3 [81, 82]. HCAR1 is activated by lactate whereas HCAR2 is activated by the ketone body 3-hydroxy-butyric acid, and HCAR3 is activated by 3-hydoxy-octanoic acid (an intermediate in β-oxidation of octanoate). Lactate binding to HCAR1 stimulates GTPγS-binding with an EC50 of 1.3 mM. However, HCAR1 cloned from tissues of mice, rats, dogs, pigs, cows, monkeys, and Zebra fish responds to physiological concentrations of lactate at 0.5–2.0 mM, as well as 10–20 mM lactate. HCAR1 is expressed in human pituitary, adipocytes, and brown adipose tissue, as well as uterine epithelia and conceptus trophectoderm of sheep [see [80]]. In tumor cells, lactate regulates expression of HCAR1 mRNA via STAT3 (signal transducer and activator of transcription factor 3) cell signaling. In non-immune cells, lactate-HCAR1 cell signaling activates protein kinase A (PKA) and ERK (extracellular signal-related kinases) pathways, and plasmacytoid dendric cells are induced to express interferon alpha (IFNA) in response to calcium mobilization and calcium-calmodulin dependent protein kinase II (CaMKII) and calcineurin (CaN) phosphatase [82]. HCAR1 protein is expressed by uterine LE and sGE, as well as trophectoderm, but not GE, myometrium from sheep on Day 17 of pregnancy, as well as all cell types of the uterus on Days 30, 70, 90, 110, and 125 of gestation [80].

HIF1A is important for the establishment and maintenance of pregnancy in mammals as conceptuses develop in a low oxygen environment and respond to changes in oxygen tension, hormones, and other molecules. For example, expression of HIF1A is upregulated by progesterone in the uterus [89, 90], while HIF2A is upregulated by estrogen [90]. In sheep, HIF1A mRNA is induced by progesterone in the endometrium and HIF2A is upregulated in response to progesterone and IFNT [89]. Lactate is a ligand for HCAR1 in mammary tumors and is designated an orphan G-protein coupled receptor [91]. Lactate interactions with HCAR1 in cancer cells [92] promote angiogenesis [91], tumor growth [93], and chemoresistance [94], and proliferation and migration of normal cells [92]. Lactate produced via glycolysis is used potentially as: 1) an energy source for mitochondrial respiration; 2) gluconeogenic precursor; and 3) cell signaling molecule at physiological concentrations of lactate from 0.5 to 20 mM and when the lactate/pyruvate ratio ranges from 10 to greater than 500 mM under conditions such as vigorous exercise and stress [81]. Lactate acting via HCAR1 in adipocytes inhibits lipolysis by decreasing mitochondrial fatty acid uptake via malonyl-CoA and carnitine palmitoyltransferase I in muscle [84].

The high lactate and low pH environment in the uterine lumen during early pregnancy is created by lactate produced by blastocysts in mice [see [95–97]]. Lactate and low pH increases expression of mRNAs for VEGFA, HCAR1, SLC2A4 (also known as glucose transporter member 4), transcription factor p65 (RELA), MCT1 and snail (SNAI1) involved in epithelial to mesenchymal cell transition in Ishikawa cells. Lactic acid also increases migration of decidualized stromal cells in uteri of mice without changing the extent of decidualization. Further, human umbilical vein endothelial cells (HUVEC) form tubes when treated with 5 mM lactic acid as evidence of an angiogenic effect of lactic acid. Garner [95] reported that mammalian blastocysts use aerobic glycolysis as do cancer cells to create a microenvironment in which the pH is low to increase angiogenesis, vascular permeability, tissue disaggregation through breakdown of the extracellular matrix associated with increases in expression of matrix metalloproteinases 1 and 2 (MMP1 and 2) from blastocysts and MMP9, transforming growth factors beta 1 and 2 (TGFB1, TGFB2), cathepsin B and hyaluronic acid. The increase in hyaluronic acid is suggested to increase hydration of the endometrium due to facilitate implantation. Also, there was a decrease in expression of tissue inhibitors of metalloproteinases (TIMPs) and an increase in NFKB in that study [95]. Lactate also increased Treg cells, conversion of macrophages from M1 (inflammatory) to M2 (anti-inflammatory) phenotypes, and expression of VEGF in macrophages. Gardner [95] suggests that post-implantation conceptuses become more dependent on glycolysis to produce lactate and maintain a low pH environment that mimics hypoxia.

Comline and Silver [98] reported concentrations (mg/100 mL) of lactic acid in fetal umbilical vein blood at 9, 5 and 1 day prepartum to be 16.7, 16.8, and 19.1, respectively, as compared to values in blood from the maternal uterine vein of 11.2, 9.8 and 9.8. In comparison, concentrations (mg/100 mL) of fructose in fetal umbilical vein blood at 9, 5, and 1 day prepartum were 74.3, 77.8, and 67.7, respectively, but not detectable in maternal blood samples. The concentrations of glucose in umbilical vein blood were 13.3, 12.7, and 19.1 mg/100 mL at 9, 5 and 1 day prepartum, respectively as compared to 57.6, 58.0, and 68.0 mg/100 mL in maternal uterine vein blood. Thus, concentrations of fructose in fetal umbilical vein blood were 5- to 6-fold greater than those for glucose in that study [98].

The literature documents that the intrauterine environment for mammalian conceptuses is hypoxic relative to normal air [95, 99–102] and that is also true for developing tumors [61]. Accordingly, the polyol pathway is active in conceptuses of ungulates such as pigs [57] and sheep [54] and even humans [32]. The polyol pathway generates fructose and the metabolism of fructose via frutolysis is sustained for conceptuses of pigs [57] and sheep [33] throughout gestation, but not for human conceptuses [32]. We speculate that species with invasive implantation and hemochorial or hemoendothelial placenta with fewer layers of tissue separating maternal and fetal blood do not require high rates of blood flow to the uterus or a well-developed allantois to serve as a reservoir of nutrients. The various species of ungulates have superficial implantation of the blastocyst/conceptus and high rates of uterine blood flow to ensure a sustained delivery of high amounts of nutrients [concentration of nutrient X uterine blood flow] for transfer to the fetal-placental vasculature, as well as a well-developed allantois in which nutrients not utilized immediately can accumulate and be recycled to ensure that the conceptus is well nourished.

Adaptation of the polyol pathway and fructolysis in the placenta has several advantages for ungulates (see Figure 5). First, the trophectoderm/chorioallantois rapidly converts available glucose to fructose that cannot be transferred to the maternal vasculature, it is a sequestered hexose sugar. Second, fructose is phosphorylated at carbon 1 to F1P that is then committed to the fructolysis pathway for further metabolism to substrates that support the pentose cycle, TCA cycle, hexosamine biosynthesis pathway, and one-carbon metabolism. Third, the fructolysis metabolic pathway is not inhibited by low pH, ATP or citrate as is the case for the hexosamine-dependent pathway for glycolysis. Fourth, the fructose in fetal blood is excreted in urine during the first 24–48 h after birth so that does not cause insulin-insensitivity [102] as piglets fail to survive on synthetic diets containing only fructose [103, 104]. Failure of newborn piglets to survive on fructose-based synthetic diets further validates the unique role of fructose and the fructolysis pathway used by fetal-placental tissue of ungulates. Further, it is now becoming apparent that lactate, a product of fructose metabolism, likely acts via its own receptor (HCAR1) to influence implantation and placentation, as well as sustain expression of HIF1A and its downstream genes such as KHK and VEGF [105–107]. On-going and future research will further expand knowledge of the roles of fructose and lactate in fetal-placental development and cancer as they represent rapidly developing tissues that share many metabolic profiles.