Gene expression analyses of TAS1R taste receptors relevant to the treatment of cardiometabolic disease.

The sweet taste receptor (STR) is a G protein-coupled receptor (GPCR) responsible for mediating cellular responses to sweet stimuli. Early evidence suggests that elements of the STR signaling system are present beyond the tongue in metabolically active tissues, where it may act as an extraoral glucose sensor. This study aimed to delineate expression of the STR in extraoral tissues using publicly available RNA-sequencing repositories. Gene expression data was mined for all genes implicated in the structure and function of the STR, and control genes including highly expressed metabolic genes in relevant tissues, other GPCRs and effector G proteins with physiological roles in metabolism, and other GPCRs with expression exclusively outside the metabolic tissues. Since the physiological role of the STR in extraoral tissues is likely related to glucose sensing, expression was then examined in diseases related to glucose-sensing impairment such as type 2 diabetes. An aggregate co-expression network was then generated to precisely determine co-expression patterns among the STR genes in these tissues. We found that STR gene expression was negligible in human pancreatic and adipose tissues, and low in intestinal tissue. Genes encoding the STR did not show significant co-expression or connectivity with other functional genes in these tissues. In addition, STR expression was higher in mouse pancreatic and adipose tissues, and equivalent to human in intestinal tissue. Our results suggest that STR expression in mice is not representative of expression in humans, and the receptor is unlikely to be a promising extraoral target in human cardiometabolic disease.

Orphan GPR146: an alternative therapeutic pathway to achieve cholesterol homeostasis?

Atherosclerosis predisposes to myriad cardiovascular complications, including myocardial infarction and stroke. Statins have revolutionised cholesterol management but they do not work for all patients, particularly those with familial hypercholesterolaemia (FH). Genome-wide association studies have linked SNPs at orphan G protein-coupled receptor 146 (GPR146) to human atherosclerosis but how GPR146 influences serum cholesterol homeostasis was only recently described. Gpr146 deletion in mice reduces serum cholesterol and atherosclerotic plaque burden, confirming GPR146 as a potential therapeutic target for managing circulating cholesterol. Critically, this effect was independent of the low-density lipoprotein receptor. While still an orphan, the activation of GPR146 by serum suggests identification of its endogenous ligand is tantalisingly close. Herein, we discuss the evidence for GPR146 inhibition as a treatment for atherosclerosis.

Characterisation of bis(4-aminoquinoline)s as α1A adrenoceptor allosteric modulators.

The development of sub-type selective α1 adrenoceptor ligands has been hampered by the high sequence similarity of the amino acids forming the orthosteric binding pocket of the three α1 adrenoceptor subtypes, along with other biogenic amine receptors. One possible approach to overcome this issue is to target allosteric sites on the α1 adrenoceptors. Previous docking studies suggested that one of the quinoline moieties of a bis(4-aminoquinoline), comprising a 9-carbon methylene linker attached via the amine groups, could interact with residues outside of the orthosteric binding site while, simultaneously, the other quinoline moiety bound within the orthosteric site. We therefore hypothesized that this compound could act in a bitopic manner, displaying both orthosteric and allosteric binding properties. To test this proposition, we investigated the allosteric activity of a series of bis(4-aminoquinoline)s with linker lengths ranging from 2 to 12 methylene units (designated C2-C12). A linear trend of increasing [3H]prazosin dissociation rate with increasing linker length between C7 and C11 was observed, confirming their action as allosteric modulators. These data suggest that the optimal linker length for the bis(4-aminoquinoline)s to occupy the allosteric site of the α1A adrenoceptor is between 7 and 11 methylene units. In addition, the ability of C9 bis(4-aminoquinoline) to modulate the activation of the α1A adrenoceptor by norepinephrine was subsequently examined, showing that C9 acts as a non-competitive antagonist. Our findings indicate that the bis(4-aminoquinolines) are acting as allosteric modulators of orthosteric ligand binding, but not efficacy, in a bitopic manner.

Critically evaluating sweet taste receptor expression and signaling through a molecular pharmacology lens.

The class C G protein-coupled sweet taste receptor (STR) is responsible for the perception of sweet-tasting molecules. Considered an obligate heterodimer, it consists of taste 1 receptor 2 and taste 1 receptor 3 subunits. Interest in the STR has steadily grown, especially since its discovery in extraoral tissues hints at a metabolic role for the receptor. It is now known that many pharmacologically exploitable binding sites exist across the extracellular and transmembrane regions of both subunits of the STR, indicative of its potential amenability to pharmacotherapeutic modulation. In this review, we briefly describe the structural characteristics and functional relevance of the STR. Then, from a molecular pharmacology perspective, we dissect the research surrounding the regulation of STR surface expression and signal transduction, in both oral and extraoral tissues, and discuss the potential for the exploitation of biased agonists for the STR. We find that despite 20 years of research into the STR, the target remains frustratingly enigmatic. Not only are the mechanisms controlling and regulating the surface expression of the STR unclear, but also research into the full repertoire of signaling partners of the STR is at present inconclusive. Critically, the influence of receptor polymorphisms (including those associated with sugar consumption) on the molecular pharmacology of the receptor remains hitherto unexplored. Finally, we provide recommendations on the reporting of reference sequence identification numbers to avoid incorrect attribution of wild-type to these biologically significant polymorphisms, which we argue may have led to some of the inconsistencies in the field.

Design, Synthesis, and Evaluation of N- and C-Terminal Protein Bioconjugates as G Protein-Coupled Receptor Agonists.

A G protein-coupled receptor (GPCR) agonist protein, thaumatin, was site-specifically conjugated at the N- or C-terminus with a fluorophore for visualization of GPCR:agonist interactions. The N-terminus was specifically conjugated using a synthetic 2-pyridinecarboxyaldehyde reagent. The interaction profiles observed for N- and C-terminal conjugates were varied; N-terminal conjugates interacted very weakly with the GPCR of interest, whereas C-terminal conjugates bound to the receptor. These chemical biology tools allow interactions of therapeutic proteins:GPCR to be monitored and visualized. The methodology used for site-specific bioconjugation represents an advance in application of 2-pyridinecarboxyaldehydes for N-terminal specific bioconjugations.

Template selection and refinement considerations for modelling aminergic GPCR-ligand complexes.

G protein-coupled receptors (GPCRs) are important targets for development of drugs for the treatment of many diseases. However, crystal structures are available for only a small fraction of these membrane bound proteins. Accurate homology models will provide opportunities for effective drug design targeting GPCRs. Recently, several serotonin receptor crystal structures were solved and needed to be evaluated as potential templates. In the first part of this work different measures of similarity in template selection were explored and methods for homology modelling, docking and refinement of aminergic GPCR-ligand complexes were developed and evaluated by comparing models of the D3-R/eticlopride complex with the crystal structure. Homology models of the three α1 adrenergic receptor subtypes and of a serotonin receptor subtype were then constructed using these methods These models were evaluated by docking a range of antagonists into them.

Homobivalent Conjugation Increases the Allosteric Effect of 9-aminoacridine at the α1-Adrenergic Receptors.

The α1-adrenergic receptors are targets for a number of cardiovascular and central nervous system conditions, but the current drugs for these receptors lack specificity to be of optimal clinical value. Allosteric modulators offer an alternative mechanism of action to traditional α1-adrenergic ligands, yet there is little information describing this drug class at the α1-adrenergic receptors. We have identified a series of 9-aminoacridine compounds that demonstrate allosteric modulation of the α1A- and α1B-adrenergic receptors. The 9-aminoacridines increase the rate of [3H]prazosin dissociation from the α1A- and α1B-adrenergic receptors and noncompetitively inhibit receptor activation by the endogenous agonist norepinephrine. The structurally similar compound, tacrine, which is a known allosteric modulator of the muscarinic receptors, is also shown to be a modulator of the α1-adrenergic receptors, which suggests a general lack of selectivity for allosteric binding sites across aminergic G protein-coupled receptor. Conjugation of two 9-aminoacridine pharmacophores, using linkers of varying length, increases the potency and efficacy of the allosteric effects of this ligand, likely through optimization of bitopic engagement of the allosteric and orthosteric binding sites of the receptor. Such a bivalent approach may provide a mechanism for fine tuning the efficacy of allosteric compounds in future drug design efforts.

Human α1-adrenoceptor subtype selectivity of substituted homobivalent 4-aminoquinolines.

A series of ring-substituted ethyl- and heptyl-linked 4-aminoquinoline dimers were synthesized and evaluated for their affinities at the 3 human α(1)-adrenoceptor (α(1)-AR) subtypes and the human serotonin 5-HT(1A)-receptor (5-HT(1A)-R). We find that the structure-specificity profiles are different for the two series at the α(1)-AR subtypes, which suggests that homobivalent 4-aminoquinolines can be developed with α(1)-AR subtype selectivity. The 8-methyl (8-Me) ethyl-linked analogue has the highest affinity for the α(1A)-AR, 7 nM, and the greatest capacity for discriminating between α(1A)-AR and α(1B)-AR (6-fold), α(1D)-AR (68-fold), and the 5-HT(1A)-R (168-fold). α(1B)-AR selectivity was observed with the 6-methyl (6-Me) derivative of the ethyl- and heptyl-linked 4-aminoquinoline dimers and the 7-methoxy (7-OMe) derivative of the heptyl-linked analogue. These substitutions result in 4- to 80-fold selectivity for α(1B)-AR over α(1A)-AR, α(1D)-AR, and 5-HT(1A)-R. In contrast, 4-aminoquinoline dimers with selectivity for α(1D)-AR are more elusive, since none studied to date has greater affinity for the α(1D)-AR over the other two α(1)-ARs. The selectivity of the 8-Me ethyl-linked 4-aminoquinoline dimer for the α(1A)-AR, and 6-Me ethyl-linked, and the 6-Me and 7-OMe heptyl-linked 4-aminoquinoline dimers for the α(1B)-AR, makes them promising leads for drug development of α(1A)-AR or α(1B)-AR subtype selective ligands with reduced 5-HT(1A)-R affinity.

An aspartate in the second extracellular loop of the α(1B) adrenoceptor regulates agonist binding.

The extracellular loops of the adrenoceptors present a potential therapeutic target in the design of highly selective adrenergic drugs. These regions are less conserved than the orthosteric binding site but have to date not been implicated in activation of adrenoceptors. A previously generated homology model identified an extracellular residue, D191, as a potential regulator of agonist binding. We have generated mutants of the α1B adrenoceptor replacing the charged aspartate, D191, as well as a potential interaction partner, K331, with uncharged alanines to observe effects on ligand binding and receptor activation. Significant 4-6 fold reductions in affinity for the endogenous agonists, epinephrine and norepinephrine were observed for receptors with the D191A mutation in the second extracellular loop. While changes in EC50 were observed, operational analysis yielded no apparent change in receptor activation. Based on these findings, we suggest that D191, in the second extracellular loop of the α1B adrenoceptor, acts as a 'point of first contact' for the receptor's endogenous agonists. Implication of the non-conserved extracellular regions of the receptor in agonist binding makes it a potential target for the design of highly selective drugs.

Expression of adrenoceptor subtypes in preterm piglet heart is different to term heart.

Preterm delivery increases the risk of inadequate systemic blood flow and hypotension, and many preterm infants fail to respond to conventional inotrope treatments. If the profile of cardiac adrenoceptor subtypes in the preterm neonate is different to that at term this may contribute to these clinical problems. This study measured mRNA expression of ß1, ß2, α1A, α2A and α2B-adrenoceptor subtypes by real time PCR in term (113d), preterm (91d) and preterm piglets (91d) exposed to maternal glucocorticoid treatment. Abundance of ß-adrenoceptor binding sites in the left ventricle was measured using saturation binding assays. Relative abundance of ß1-adrenoceptor mRNA in untreated preterm hearts was ∼50% of term abundance in both left and right ventricles (P<0.001). Trends in receptor binding site density measurements supported this observation (P = 0.07). Glucocorticoid exposure increased ß1-adrenoceptor mRNA levels in the right ventricle of preterm hearts (P = 0.008) but did not alter expression in the left ventricle (P>0.1). Relative abundance of α1A-adrenoceptor mRNA was the same in preterm and term piglet hearts (P = >0.1) but was reduced by maternal glucocorticoid treatment (P<0.01); α2A-adrenoceptor mRNA abundance was higher in untreated and glucocorticoid exposed preterm piglet hearts than in term piglets (P<0.001). There was no difference between male and female piglets in mRNA abundance of any of the genes studied. In conclusion, there is reduced mRNA abundance of ß1-adrenoceptors in the preterm pig heart. If this lower expression of ß-adrenoceptors occurs in human preterm infants, it could explain their poor cardiovascular function and their frequent failure to respond to commonly used inotropes.

5-HT1A receptor pharmacophores to screen for off-target activity of α1-adrenoceptor antagonists.

The α1-adrenoceptors (α1-ARs), in particular the α1A-AR subtype, are current therapeutic targets of choice for the treatment of urogenital conditions, such as benign prostatic hyperplasia (BPH). Due to the similarity between the transmembrane domains of the α1-AR subtypes, and the serotonin receptor subtype 1A (5-HT1A-R), currently used α1-AR subtype-selective drugs to treat BPH display considerable off-target affinity for the 5-HT1A-R, leading to side effects. We describe the construction and validation of pharmacophores for 5-HT1A-R agonists and antagonists. Through the structural diversity of the training sets used in their development, these pharmacophores define the properties of a compound needed to bind to 5-HT1A receptors. Using these and previously published pharmacophores in virtual screening and profiling, we have identified unique chemical compounds (hits) that fit the requirements to bind to our target, the α1A-AR, selectively over the off-target, the 5-HT1A-R. Selected hits have been obtained and their affinities for α1A-AR, α1B-AR and 5-HT1A-R determined in radioligand binding assays, using membrane preparations which contain human receptors expressed individually. Three of the tested hits demonstrate statistically significant selectivity for α1A-AR over 5-HT1A-R. All seven tested hits bind to α1A-AR, with two compounds displaying K i values below 1 µM, and a further two K i values of around 10 µM. The insights and knowledge gained through the development of the new 5-HT1A-R pharmacophores will greatly aid in the design and synthesis of derivatives of our lead compound, and allow the generation of more efficacious and selective ligands.

α1-Adrenoceptor and serotonin 5-HT(1A) receptor affinity of homobivalent 4-aminoquinoline compounds: an investigation of the effect of linker length.

α1-adrenoceptor (α1-AR) subtype-selective ligands lacking off-target affinity for the 5-HT(1A) receptor (5-HT(1A)-R) will provide therapeutic benefits in the treatment of urogenital conditions such as benign prostatic hyperplasia. In this study we determined the affinity of 4-aminoquinoline and eleven homobivalent 4-aminoquinoline ligands (diquinolines) with alkane linkers of 2-12 atoms (C2-C12) for α(1A), α(1B) and α(1D)-ARs and the 5-HT(1A)-R. These ligands are α(1A)-AR antagonists with nanomolar affinity for α(1A) and α(1B)-ARs. They display linker-length dependent selectivity for α(1A/B)-ARs over α(1D)-AR and the 5-HT(1A)-R. The C2 diquinoline has the highest affinity for α1A-AR (pKi 7.60±0.26) and greater than 30-fold and 600-fold selectivity for α(1A)-AR over α(1D)-AR and 5-HT(1A)-R respectively. A decrease in affinity for α1-ARs is observed as the linker length increases, reaching a nadir at 5 (α(1A/1B)-ARs) or 6 (α(1D)-AR) atoms; after which affinity increases as the linker is lengthened, peaking at 9 (α(1A/1B/1D)-ARs) or 8 (5-HT(1A)-R) atoms. Docking studies suggest that 4-aminoquinoline and C2 bind within the orthosteric binding site, while for C9 one end is situated within the orthosteric binding pocket, while the other 4-aminoquinoline moiety interacts with the extracellular surface. The limited α(1D)-AR and 5-HT(1A)-R affinity of these compounds makes them promising leads for future drug development of α(1A)-AR selective ligands without α(1D)-AR and the 5-HT(1A)-R off-target activity.

A novel structural framework for α(1A/D)-adrenoceptor selective antagonists identified using subtype selective pharmacophores.

In this study four and five-feature pharmacophores for selective antagonists at each of the three α(1)-adrenoceptor (AR) subtypes were used to identify novel α(1)-AR subtype selective compounds in the National Cancer Institute and Tripos LeadQuest databases. 12 compounds were selected, based on diversity of structure, predicted high affinity and selectivity at the α(1D)- subtype compared to α(1A)- and α(1B)-ARs. 9 out of 12 of the tested compounds displayed affinity at the α(1A) and α(1D) -AR subtypes and 6 displayed affinity at all three α(1)-AR subtypes, no α(1B)-AR selective compounds were identified. 8 of the 9 compounds with α(1)-AR affinity were antagonists and one compound displayed partial agonist characteristics. This virtual screening has successfully identified an α(1A/D)-AR selective antagonist, with low µM affinity with a novel structural scaffold of a an isoquinoline fused three-ring system and good lead-like qualities ideal for further drug development.

Glucocorticoid exposure and tissue gene expression of 11beta HSD-1, 11beta HSD-2, and glucocorticoid receptor in a porcine model of differential fetal growth.

Glucocorticoids play a critical role in fetal development, but inappropriate exposure is associated with reduced fetal growth. We investigated cortisol exposure and supply in a porcine model of differential fetal growth. This model compares the smallest fetus of a litter with an average-sized sibling at three stages of gestation. At day 45, small fetuses had reduced plasma cortisol (16.8 +/- 3.4 ng/ml) relative to average fetuses (34.4 +/- 3.4 ng/ml, P < 0.001). At day 65 levels had reduced in small and average fetuses to similar concentrations (5.7 +/- 1.0 vs 4.8 +/- 0.5 ng/ml, P = 0.128). By day 100, elevated levels were found in small fetuses (10.7 +/- 1.5 vs 7.6 +/- 0.7 ng/ml, P < 0.001). Maternal plasma cortisol was unchanged over gestation (day 45, 56.7 +/- 21.6 ng/ml; day 65, 57.8 +/- 14.4 ng/ml; day 100, 55.7 +/- 6.5 ng/ml). We examined the cause of altered cortisol by investigating the fetal hypothalamic-pituitary-adrenal axis through the measurement of adrenocorticotropic hormone and assessing exposure to maternal cortisol by quantifying placental 11beta-hydroxysteroid dehydrogenase-isoform 2 (11beta HSD-2) gene expression. These data suggest that altered cortisol supply was of fetal origin. We examined organ glucocorticoid (GC) metabolism by the measurement of GC receptor (GR) and 11beta-hydroxysteroid dehydrogenase-isoform 1 (11beta HSD-1) gene expression. We found that fetal organs have different temporal patterns of 11beta HSD-1 and GR expression, with the liver particularly sensitive to cortisol in late gestation. This study examines GC exposure in naturally occurring differential growth and simultaneously explores tissue GC sensitivity and handling, at three key stages of gestation.

Transgenic alpha1A-adrenergic activation limits post-infarct ventricular remodeling and dysfunction and improves survival.

OBJECTIVE: Myocardial contractility is enhanced in transgenic (TG) mice with cardiac-restricted overexpression of the alpha1A-adrenergic receptors (alpha1A-AR). We tested the hypothesis that this enhanced inotropy protects against dysfunction and remodeling after myocardial infarction (MI). METHODS: We subjected alpha1A-TG and non-TG mice (NTG) to MI and determined changes in left ventricular (LV) function and diastolic dimension (LVDd) by echocardiography prior to and at 1, 3, 7, 12 and 15 weeks thereafter. RESULTS: Although infarct size was similar in the NTG and alpha1A-TG groups (32+/-2 vs. 29+/-2% of LV, P=NS), mortality due to heart failure was lower after MI in the alpha1A-TG (37%, n=39) than that in the NTG animals (63%, n=56, P=0.026). NTG and alpha1A-TG mice showed similar reductions in LV fractional shortening (FS) and increases in LVDd at week-1 after MI. However, whereas NTG mice showed continuous deterioration over a 15-week period after MI in FS (fell by 40%, from 30+/-2 to 18+/-1%, P<0.01) and LVDd (increased by 24%, from 4.2+/-0.1 to 5.2+/-0.1 mm, P<0.01), the changes in both FS (fell by 14%, from 42+/-2 to 36+/-2%) and LVDd (increased by 8%, from 3.8+/-0.1 to 4.1+/-0.1 mm, both changes P<0.01 vs. NTG) were significantly less severe in the alpha1A-TG mice and did not progress after 3 weeks. At 15 weeks after MI, LV catheterization revealed better preservation of dP/dtmax in the alpha1A-TG vs. NTG mice (7270+/-324, vs. 5938+/-372 mmHg/s, P<0.05). CONCLUSION: Enhanced inotropy resulting from transgenic overexpression of alpha1A-AR is well maintained chronically after MI and limits echocardiography-determined LV remodeling, preserves function, and reduces acute heart failure death.

The alpha(1D)-adrenergic receptor: cinderella or ugly stepsister.

This Perspective focuses on the alpha(1D)-adrenergic receptor (AR), the often neglected sibling of the alpha(1)-AR family. This neglect is due in part to its poor cell-surface expression. However, it has recently been shown that dimerization of the alpha(1D)-AR with either the alpha(1B)-AR or the beta(2)-AR increases alpha(1D)-AR cell-surface expression, and in this issue of Molecular Pharmacology, Hague et al. (p. 45) demonstrate that dimerization of the alpha(1D)-AR with the alpha(1B)-AR not only leads to increased cell-surface expression but also results in the formation of a novel functional entity.

The effect of fetal pig size and stage of gestation on tissue fatty acid metabolism and profile.

The fetus requires an adequate supply of fatty acids for optimum growth and development. It has been hypothesized that reduced activity of enzymes of fatty acid metabolism could contribute to inadequate fetal growth. In a porcine model of differential fetal growth we examined heart and liver fatty acid synthase, delta5-desaturase and delta6-desaturase gene expression and measured hepatic fatty acid profile to assess long-chain polyunsaturated fatty acid status. On gestation days 45, 65 and 100 sows were killed and tissues extracted from an average-sized fetus and the smallest fetus from each litter. As early as day 45, considerable hepatic delta5- and delta6-desaturase was detected, and this expression significantly increased as gestation progressed. In contrast, cardiac desaturase expression remained stable with time. Fatty acid synthase expression was greatest at day 65 in the liver, but was not expressed in the heart. Overall, the smallest fetus did not exhibit reduced tissue delta5- or delta6-desaturase expression or compromised polyunsaturated fatty acid status at any stage. In fact, small fetuses expressed more cardiac delta5-desaturase than their average-sized siblings, possibly in response to a stress to the heart. It is clear from this study that fatty acid metabolism changes markedly as gestation progresses, and reduced fatty acid supply does not cause inadequate growth in this porcine model of fetal development.

Genetic enhancement of ventricular contractility protects against pressure-overload-induced cardiac dysfunction.

In response to pressure-overload, cardiac function deteriorates and may even progress to fulminant heart failure and death. Here we questioned if genetic enhancement of left ventricular (LV) contractility protects against pressure-overload. Transgenic (TG) mice with cardiac-restricted overexpression (66-fold) of the alpha(1A)-adrenergic receptor (alpha(1A)-AR) and their non-TG (NTG) littermates, were subjected to transverse aorta constriction (TAC)-induced pressure-overload for 12 weeks. TAC-induced hypertrophy was similar in the NTG and TG mice but the TG mice were less likely to die of heart failure compared to the non-TG animals (P <0.05). The hypercontractile phenotype of the TG mice was maintained over the 12-week period following TAC with LV fractional shortening being significantly greater than in the NTG mice (42+/-2 vs 29+/-1%, P <0.01). In the TG animals, 11-week beta-AR-blockade with atenolol neither induced hypertrophy nor suppressed the hypercontractile phenotype. The hypertrophic response to pressure-overload was not altered by cardiac alpha(1A)-AR overexpression. Moreover, the inotropic phenotype of alpha(1A)-AR overexpression was well maintained under conditions of pressure overload. Although the functional decline in contractility with pressure overload was similar in the TG and NTG animals, given that contractility was higher before TAC in the TG mice, their LV function was better preserved and heart failure deaths were fewer after induction of pressure overload.

Placental transport of leucine in a porcine model of low birth weight.

Low birth weight is a major factor in neonatal morbidity and mortality in humans and domestic species and is a predictor of physiological disorders in adulthood. This study utilised the naturally occurring variation in pig fetal size within a uterus to test the hypothesis that placental amino acid transport capability is associated with fetal growth. Leucine uptake by trophoblast vesicles prepared from placentas supplying an average-sized fetus and the smallest fetus in the uterus was assessed. On days 45 and 65 of gestation, uptake of leucine by the porcine placenta was predominantly sodium independent and was inhibited by the non-metabolised leucine analogue 2-amino-2-norbornane-carboxylic acid, indicating that uptake occurs via system L. By day 100 the uptake of leucine by placentas supplying average-sized fetuses had changed from being predominantly sodium independent to involving both sodium-dependent (system B0) and -independent (system L) pathways. This change was not seen in placentas supplying the smallest fetus, which continued to display predominantly sodium-independent uptake. In conclusion, these data show gestational- and fetal size-dependent changes in the transport of leucine across the porcine placenta.