Renal maturation occurs post-natally in many species and reabsorption capacity at birth can vary substantially from the mature kidney. However, little is known regarding the maturation of amino acid transport mechanisms, despite the well-known physiological state of developmental iminoglycinuria. Commonly seen during early infancy, developmental iminoglycinuria is a transient version of the persistent inherited form of the disorder, referred to as iminoglycinuria, and manifests as a urinary hyperexcretion of proline, hydroxyproline and glycine. The transporters involved in developmental iminoglycinuria and their involvement in the improvement of renal reabsorption capacity remain unknown. qPCR (quantitative real-time PCR) and Western blot analysis in developing mouse kidney revealed that the expression of Slc6a18, Slc6a19, Slc6a20a and Slc36a2 was lower at birth (approx. 3.4-, 5.0-, 2.4- and 3.0-fold less than adult kidney by qPCR respectively) and increased during development. Furthermore, immunofluorescence confocal microscopy demonstrated the absence of apical expression of Slc6a18, Slc6a19, Slc6a20a and the auxiliary protein collectrin in kidneys of mice at birth. This correlated with the detection of iminoglycinuria during the first week of life. Iminoglycinuria subsided (proline reduction preceded glycine) in the second week of life, which correlated with an increase in the expression of Slc6a19 and Slc6a20a. Mice achieved an adult imino acid and glycine excretion profile by the fourth week, at which time the expression level of all transporters was comparable with adult mice. In conclusion, these results demonstrate the delayed expression and maturation of Slc6a18, Slc6a19, Slc6a20a and Slc36a2 in neonatal mice and thus the molecular mechanism of developmental iminoglycinuria.
- amino acid transporter
- developmental iminoglycinuria
- kidney development
It has been known for half a century that the epithelial uptake of amino acids is controlled by specific systems of membrane-spanning proteins, termed solute carriers or transporters [1–5]. The mammalian intestine and kidney possess similar but distinct sets of apical and basolateral transporters . Their importance, as well as the mechanisms by which they function, has been highlighted through the study of inherited aminoacidurias which disrupt normal intestinal absorption and renal reabsorption of amino acids [7–10]. Iminoglycinuria [OMIM (Online Mendelian Inheritance in Man) #242600] is one of the more common forms of inherited aminoaciduria in humans , although precise estimates are confounded by a lack of accurate adult data. This autosomal-recessive disorder affects renal, and in some cases intestinal, transport and is diagnosed by an increase in the urinary excretion of the imino acids proline and hydroxyproline, as well as the amino acid glycine [12,13].
In the kidney, proline and glycine transport is mediated by four transporters, namely the proline and glycine transporter SLC36A2 (PAT2) (where SLC36 is solute carrier family 36), the glycine transporter SLC6A18 (B0AT3 or XT2) (where SLC6 is solute carrier family 6), the neutral amino acid transporter SLC6A19 (B0AT1), which is affected in Hartnup disorder [7,9], and the proline-specific transporter SLC6A20 (IMINO or SIT1). A high degree of similarity exists between human and mouse orthologues of these transporters. However, the mouse genome possesses two genes corresponding to human SLC6A20; Slc6a20a is a functional gene, whereas no function has been observed for Slc6a20b . The role of SLC36A1 (PAT1) in the kidney remains unresolved. This transporter has been shown to localize to the apical membrane of intestinal enterocytes and accepts both imino acids and glycine . However, the distribution of SLC36A1 within the kidney is unknown. Interestingly, several members of the SLC6 family, including Slc6a19 [16–18] and Slc6a18 , require auxiliary proteins such as Ace2 (angiotensin-converting enzyme 2) and collectrin for optimal functional expression at the cell surface. The significance of these interactions is evident from the aminoaciduria and altered expression of transporters, including Slc6a18, Slc6a19 and Slc6a20a, in Ace2- and Tmem27 (transmembrane protein 27; collectrin)-null mice [18–20]. The need for these cofactors has also been emphasized by the phenotypic heterogeneity observed in cases of Hartnup disorder where mutated SLC6A19 differentially interacts with ACE2 and collectrin within kidney and intestine [16,20].
We have recently shown that inactivation or reduced function of SLC36A2 is the predominant determinant of the inherited iminoglycinuria phenotype in humans . Mutations that co-segregate with the disorder in additional transporters, including SLC6A19, SLC6A20 and SLC6A18, were also identified. However, it is also common to observe temporarily increased levels of the imino acids and glycine in the urine of normal infants . The term iminoglycinuria refers to both the persistent inherited abnormality associated with transporter mutations as well as the neonatal developmental phenotype that is the subject of the present paper. To distinguish these, ‘developmental iminoglycinuria’ is used throughout to refer to the latter only.
Morphological and functional maturation of the mammalian kidney occurs post-birth in many species, including humans [22,23], dogs [24,25], pigs  and rodents [27,28]. Developmental iminoglycinuria is a feature of mammalian renal ontogeny  and typically resolves within the first 12 months, 2 months and 2 weeks post-birth in humans , dogs [29,31] and rodents [32–35] respectively. Although previous studies have reported evidence that developmental iminoglycinuria results from ‘delayed maturation’ of transport systems for proline, hydroxyproline and glycine [12,36–38], the specific transporters involved have not yet been identified.
A complex set of mutations inherited in multiple transporters has only recently been identified by our group, thus explaining the molecular pathogenesis of inherited iminoglycinuria . In light of this recent discovery, a reasonable hypothesis is that delayed expression of these same transporters during early infancy is responsible for developmental iminoglycinuria. To address this, we examined the renal expression of Slc6a18, Slc6a19, Slc6a20a, Slc36a1 and Slc36a2, as well as the auxiliary proteins, collectrin and Ace2, during mouse development. The results presented herein demonstrate that excessive neonatal loss of imino acids and glycine results from the delayed maturation of Slc6a18, Slc6a19, Slc6a20a and Slc36a2, and also defines the molecular mechanisms of renal amino acid transport ontogeny.
C57Bl/6J mice obtained from the Animal Resources Centre were housed under specific pathogen-free conditions according to institutional guidelines. Mice were provided with meat-free irradiated rat and mouse feed (Gordon's Speciality Stockfeeds). Animal protocols were approved by the University of Sydney Animal Ethics Committee (Approval No. K75/8-2006/3/4425). Timed matings were undertaken in order to obtain three mice at each time point, which included 0, 7, 14, 28, 42 and 84 days post-birth. Mice up to 2 weeks of age were decapitated prior to organ retrieval, whereas mice at all other time points were killed using CO2 asphyxiation.
RNA extraction and RT (reverse transcription)
Whole excised kidneys were rinsed in 0.01% diethylpyrocarbonate then homogenized in TRIzol® reagent (Invitrogen). Total RNA was extracted and treated with DNase I (Invitrogen) and RNase H (New England BioLabs). Either 1 μg or 0.5 μg was used as a template for RT using SuperScript III enzyme (Invitrogen) and oligo(dT) (GE Healthcare). Reverse transcriptase was omitted in negative controls (−RT).
qPCR (quantitative real-time PCR) analysis
Forward and reverse qPCR primers were designed against RefSeq sequences using Primer Express v.3.0 software (Applied Biosystems) for Slc6a18 (GenBank® accession number NM_001040692; 5′-CCCACCCAGGAACAATTGTG-3′ and 5′-TTCCATAGTCATTGGATGCCTTGA-3′), Slc6a19 (GenBank® accession number NM_028878; 5′-GGGCCTGTACTACAACACCATCA-3′ and 5′-CCACATAGCCTGTCTGGTTCTG-3′), Slc6a20a (GenBank® accession number NM_139142; 5′-CGGGAACACTGCAGTACCAA-3′ and 5′-TCTTGAGGCGATTCCTGATGA-3′), Slc36a2 (GenBank® accession number NM_153170; 5′-CAAAGCCAGCATCACTCTTAACC-3′ and 5′-CACAGGATACCGACGACATACAG-3′) and Gapdh (glyceraldehyde-3-phosphate dehydrogenase; GenBank® accession number NM_008084; 5′-TGCACCACCAACTGCTTAGC-3′ and 5′-GGCATGGACTGTGGTCATGAG-3′). PCRs were performed using a RotorGene 3000 (Corbett Life Sciences) and the HotStarTaq Master Mix Kit (Qiagen) with SYBR Green chemistry. All cDNA was diluted 1:5. qPCR on kidney cDNA (+RT) from mice at each time point was performed in triplicate. Negative controls (−RT) and no-template controls were performed in duplicate.
To quantitatively compare gene expression in developing and adult kidney, triplicate Ct values from each biological replicate were averaged and used to calculate the ΔΔCt value for that time point. Ct values were normalized to the expression of Gapdh (verified using GeNorm; http://medgen.ugent.be/~jvdesomp/genorm/) and expressed relative to adult kidney, as 2−ΔΔCt. Data were drawn graphically using Prism 5 (GraphPad Software).
Western blot analysis
BBMVs (brush-border membrane vesicles) were isolated from kidneys using the one-step protocol as described previously . Proteins (7.5 μg/lane) were resolved using SDS/PAGE prior to Western blotting. Custom-made antibodies specific for transporters (all from Pineda Antibody Service, Berlin, Germany) were raised against either the stated immunogenic peptide or an N-terminal GST (glutathione transferase)-fusion protein and included 1:3000-diluted rabbit anti-Slc6a19, 1:200-diluted rabbit anti-Slc6a18 (C-YKQRWKATHLESGLKLQESRG), 1:200-diluted rabbit anti-Slc6a20a (C-VALGTFIRNRLKRGGSAPVA) and 1:200-diluted rabbit anti-Slc36a2 (C-ESAKKLENKDSTFLDESPSES). Auxiliary proteins and β-actin (used as a loading control) were detected with 1:3000-diluted mouse anti-collectrin (Alexis Biochemicals), 1:4000-diluted rabbit anti-Ace2 (Abcam) and 1:2000-diluted rabbit anti-β-actin (Abcam). Blots were probed with either HRP (horseradish peroxidase)-conjugated goat anti-(rabbit IgG) or sheep anti-(mouse IgG) antibodies (GE Healthcare), and were stripped and re-probed for each transporter due to the limited amounts of sample available. Oocytes were injected with 5 ng of Slc6a18 cRNA and 2 ng of Tmem27 (collectrin) cRNA. Total membrane protein isolation and surface biotinylation was carried out as described previously . Conditions for immunodetection were the same as for BBMVs.
The specificity of the custom-made primary antibodies specific for transporters was assessed in both the Xenopus laevis oocyte expression system and in BBMV preparations. For BBMV tests, proteins were separated by SDS/PAGE and Western blotting was performed using diluted antibody alone or diluted antibody that had been pre-adsorbed with the corresponding immunizing peptide. A concentration of 2 μg/ml was used for the immunogenic peptide. The Slc6a19 antibody was raised against a GST-fusion protein, as described previously , and the collectrin and Ace2 antibodies were from a commercial source for which no peptide was available. For oocyte tests, membrane or surface proteins isolated from oocytes that had been injected with transporter cRNA and non-injected oocytes were used. Proteins were separated by SDS/PAGE and Western blotting was performed with custom-made antibodies as described above.
Immunofluorescence confocal microscopy
Tissues to be stained for Slc6a18, Slc6a19, Slc36a1, Slc36a2 or collectrin underwent fixation and embedding as described previously . Tissues to be stained with the anti-Slc6a20a antibody were fixed in 4% (v/v) paraformaldehyde at 4 °C, cryo-protected overnight in a 30% (w/v) sucrose solution, snap-frozen in Optimal Cutting Temperature compound (ProSciTech) and 6 μm sections collected on to poly-L-lysine-coated slides. Immunofluorescence of tissue sections was performed as described previously . Polyclonal primary antibodies included 1:200-diluted rabbit anti-Slc6a18, 1:200-diluted chicken anti-Slc6a19 (custom antibody from Aves Laboratories; immunizing peptide CZ-DPNYEEFPKSQK), 1:5-diluted sheep anti-Slc6a20a (custom antibody from the Institute of Medical and Veterinary Science, Adelaide, SA, Australia; immunizing peptide H-CIRNRLKRGGSAPVA-OH ), 1:400-diluted rabbit anti-Slc36a1 (custom-made antibody from Pineda Antibody Service; immunizing peptide STQRLRNEDYHDYSSTD-C) 1:400-diluted rabbit anti-Slc36a2, 1:200-diltued rabbit anti-Ace2 and 1:200-diluted mouse anti-collectrin. Secondary antibodies consisted of 1:200-diluted Alexa Fluor® 488-coupled goat anti-(rabbit IgG) (Invitrogen), 1:500-diluted Alexa Fluor® 568-coupled donkey anti-(sheep IgG) (Invitrogen), 1:500-diluted Alexa Fluor® 594-coupled goat anti-(chicken IgG) (Invitrogen), 1:500-diluted Alexa Fluor® 546-coupled goat anti-(chicken IgG) (Invitrogen), 1:200-diluted FITC-coupled goat anti-(mouse IgG) (Chemicon), 1:400-diluted streptavidin–Alexa Fluor® 488 (Invitrogen) and 1:500-diluted streptavidin–Alexa Fluor® 594 (Invitrogen). Nuclei were detected with 1:1000-diluted DAPI (4′,6-diamidino-2-phenylindole).
Primary antibody specificity and the level of non-specific binding was assessed by pre-adsorption of the primary antibody with the immunizing peptide or staining with the pre-immune serum of the host animal. Digital images were obtained using an SP5 confocal microscope (Leica Microsystems) and processed identically using Canvas X (ACD Systems International) to enhance brightness and contrast.
Urine collected from mice was diluted 1:4 and de-proteinized with 10% (v/v) sulfosalicyclic acid (Sigma–Aldrich) containing 125 μM norleucine internal standard (Sigma–Aldrich). Pre-column derivatization of samples was accomplished using the AccQFluor reagent kit (Waters). Chromatography was performed with gradient elution on a CLASS-VP HPLC (Shimadzu) equipped with a 3.9 mm×150 mm Nova-Pak C18 AccQTag™ amino acid analysis column (Waters) at 37 °C. Eluents consisted of 0.14 M sodium acetate buffer (pH 5.2), 100% acetonitrile and HPLC-grade (18 MOhm/cm) water. Product detection was accomplished by fluorescence with excitation at 250 nm and emission at 395 nm. Data were analysed using CLASS-VP Software Version 6.14 SP2 (Shimadzu), and the concentration of imino and amino acids in the urine of developing mice was expressed relative to the average adult excretion in μmol/ml (hydroxyproline, 0.25; proline, 1.75; leucine, 0.19; and glycine, 0.22).
Oocyte flux and electrophysiological studies
X. laevis oocyte isolation and maintenance have been described in detail previously . Use of Slc6a18 , Ace2  and Tmem27 (collectrin)  have also been described previously. For expression studies, pGEM-He-Juel plasmids containing Slc6a18, Tmem27 (collectrin) and Ace2 were linearized with NotI or SalI restriction enzymes and transcribed in vitro using the T7 mMessage mMachine Kit (Ambion). Oocytes were each injected with 5 ng of cRNA encoding Slc6a18 and/or 5 ng of cRNA encoding Ace2 or 2 ng of Tmem27 (collectrin). Amino-acid-induced currents were analysed by two-electrode voltage clamp recordings (all equipment from Axon Instruments). Oocytes with a membrane potential of less than −30 mV were chosen. Once a stable membrane potential was reached under current clamp conditions, the amplifier was switched to voltage clamp mode to maintain −50 mV. Oocytes were superfused with ND96 buffer containing substrates as indicated. The generated current by the transporter at a given membrane potential was calculated as the difference of the currents measured in the presence and the absence of substrate. Substrate-induced currents showed a significant ‘run-down’ upon repeated superfusions. As a result, alanine-induced currents were repeated between other substrates and the currents were normalized to the corresponding alanine current.
Changes in gene expression of imino acid and glycine transporters in mouse kidney during development
To determine whether imino acid and glycine transporters exhibit developmental changes, the expression levels of Slc6a18, Slc6a19, Slc6a20a and Slc36a2 were measured in whole kidneys of mice at various ages. For all transporters examined, the expression level at birth (0 days) was lower than in the adult (84 days; Figure 1A). During the first week of life, Slc6a19 and Slc6a20a expression remained approximately half that of adult kidney. In contrast, expression of Slc36a2 and Slc6a18 increased by 5.3- and 6.6-fold respectively, between 0 and 7 days of age. Slc6a18 reached levels comparable with adult kidney by the end of the first week and continued to rise during the second week of life to levels approx. 2.5-fold greater than expression in the adult kidney (Figure 1A). Slc36a2 levels also increased during the first week, reaching levels approx. 2-fold higher than adult, before decreasing to 20% of adult expression by 14 days.
Biodistribution and developmental changes in protein expression of transporters and auxiliary proteins in mouse kidney
To determine whether the low mRNA levels of transporters observed during development were reflected in protein expression levels, SDS/PAGE and Western blotting was performed on BBMVs from each developmental stage. Similar to mRNA expression levels, the amount of protein for all transporters examined was found to increase during the first weeks of life (Figure 1B). Slc6a18 and Slc6a19, together with their auxiliary protein collectrin, were barely detectable at day 0, reaching maximum expression after the third week of life. In contrast, their alternate trafficking subunit, Ace2, was consistently expressed from day 0 onwards. Slc6a20a exhibited a higher molecular-mass band comparable with that of adult kidney at day 0 and reached maximal expression by day 21. A similar pattern was also observed for Slc36a2, with a band of lower molecular mass appearing by 7 days post-birth, whereas the band at day 0 resembled that of adult kidney. Some proteins generated double bands, which are likely to represent post-translational modifications. The specificity of the custom-designed antibodies was confirmed in lysates from injected and non-injected oocytes, as well as through Western blotting of vesicle protein with antibodies that had been pre-adsorbed with their corresponding peptides (Supplementary Figure S1 at http://www.BiochemJ.org/bj/428/bj4280397add.htm). The mature proteins of Slc6a18, Slc6a19 and Slc6a20a have a molecular mass of approx. 62 kDa. In addition, it is important to note that, although the intensity of the bands reflects the relative amount of protein at each time point for individual transporters, they do not provide an absolute measure of transporter abundance, owing to the different affinities and specificities of each antibody.
To confirm the quantitative Western blotting results and to examine the biodistribution and subcellular localization of the transporters, confocal immunofluorescence microscopy was performed in mouse kidney from different developmental stages. In adult mouse kidney, Slc6a19 expression was observed on the apical membrane of S1 segments of proximal tubules (Figure 2A, arrow). However, cytoplasmic Slc6a19 staining throughout the S3 proximal tubule segment was also observed (Supplementary Figure S2 at http://www.BiochemJ.org/bj/428/bj4280397add.htm). In contrast, Slc6a18 was expressed on the apical membrane of the S2–S3 segments (Figure 2B, arrow 1), and Slc6a20a was expressed on the apical membrane of all segments (S1–S3) of kidney proximal tubules (Figure 2C, arrow), consistent with and extending the results reported previously . Slc36a2 was expressed on the apical membrane along the length of the proximal tubule (S1–S3; Figure 2D, arrow 1), as well as basolaterally in a portion of the distal nephron (Figure 2D, arrow 2, inset). In contrast with the aforementioned transporters, Slc36a1 did not exhibit an apical expression pattern. Instead, expression of this transporter was observed in sub-apical compartments of cells lining the S1 segments of proximal tubules (Figure 2E, arrow). Immunofluorescence of collectrin in adult mouse kidney (Supplementary Figure S3 at http://www.BiochemJ.org/bj/428/bj4280397add.htm) confirmed the results of previous studies [18,19] and revealed a gradient of apical expression along the entire length of proximal tubules, with the highest expression in S1 segments. Collectrin was co-localized with Slc6a18 in S2–S3 (Supplementary Figure S3A) and with Slc6a19 in S1 segments of proximal tubules (Supplementary Figure S3B).
Expression of Slc6a18, Slc6a19 and Slc6a20a was absent at birth (Figures 3Ai, 3Bi and 3Ci) and Slc36a2 exhibited low-intensity apical staining (Figure 3Di, arrow). At 7 days postbirth, all transporters were expressed, but to varying degrees. Almost all Slc6a19 expression was cytoplasmic at 7 days (Figure 3Bii, arrow 1 and inset), with a small proportion of tubules displaying faint apical staining (Figure 3Bii, arrow 2). Slc6a18, Slc6a20a and Slc36a2 were apically expressed at levels comparable with adult in kidney by 7 days (Figures 3Aii, 3Cii and 3Dii, arrows and insets). In contrast, apical expression of Slc6a19 did not reach levels equivalent to adult kidney until the second week of life (Figure 3Biii, arrow and inset).
In agreement with the Western blotting results, the faint apical expression of Ace2 that was present on day 0 (Figure 4Ai) was observed to increase during the first 2 weeks of life (Figures 4Aii and 4Aiii) before decreasing in intensity towards adulthood (Figure 4Aiv). In contrast, collectrin exhibited a punctate cytoplasmic expression pattern at birth (Figure 4Bi) that disappeared by day 7 (Figure 4Bii). This was replaced by low-intensity apical expression by the second week of life that intensified substantially as the kidney continued to develop (Figures 4Biii and 4Biv).
Urinary excretion of imino and amino acids in mice during development
To determine whether developmentally regulated changes in transporter expression levels resulted in altered urinary phenotypes, urinary biochemical profiles from mice of varying ages were analysed using HPLC. Developmental iminoglycinuria was demonstrated in all mice 0–7 days post-birth (Figure 5). Both proline and hydroxyproline excretion increased substantially during the first week of life, reaching levels approx. 8- and 25-fold higher respectively, compared with adulthood (Figures 5A and 5B). However, this diminished to levels comparable with adult mice by the second week for proline and the fourth week for hydroxyproline. Glycine also exhibited higher excretion values in neonates and reached amounts approx. 12-fold higher than adult mice in the first week. By the fourth week of life, urinary glycine excretion had declined to adult levels (Figure 5C), and an adult amino acid excretion profile for the imino acids and glycine was achieved.
Consistent with the delayed expression of the general neutral amino acid transporter Slc6a19, leucine was found to exhibit a similar excretion pattern to the imino acids and glycine, reaching levels approx. 8-fold higher than adult mice during the first neonatal week. This hyperexcretion declined substantially, along with proline and hydroxyproline, during the first neonatal week, and adult levels were achieved after the sixth week of life (Figure 5D).
Transport activity of Slc6a18
We have demonstrated previously the functional expression of Slc36a2 , Slc6a20a  and Slc6a19 . To demonstrate the transport activity of Slc6a18, the transporter was co-expressed with its auxiliary proteins in X. laevis oocytes, since Slc6a18 expressed alone does not exhibit transport activity [21,43]. In contrast with the results reported by Singer et al. , we found that the transport activity of Slc6a18 increased dramatically following co-expression with either collectrin or Ace2 (Figure 6A). The increase was due to a stabilization of the Slc6a18 protein and promotion of plasma-membrane expression in the presence of collectrin (Figure 6B). Plasma membrane expression of Slc6a18 was not observed in the absence of collectrin. The Slc6a18 protein did not appear to have an effect on the surface expression of collectrin. To investigate the substrate specificity of Slc6a18, we examined the size of substrate-induced inward currents caused by Na+-amino acid co-transport (Figure 6C) in oocytes co-expressing Slc6a18 and collectrin. The transporter was found to prefer glycine and alanine, but showed some transport activity for most neutral amino acids excluding proline (Figure 6D). Alanine showed a slightly higher maximal velocity (Figure 6E; determined at a substrate concentration of 10 mM) and a higher affinity for the transporter (0.9±0.4 mM) compared with glycine (2.3±0.4 mM).
The mammalian kidney undergoes significant morphological and functional changes post-natally that are necessary to support the growing body and reach physiological maturation. However, despite having been studied extensively, many aspects of kidney development remain poorly understood and the maturation of amino acid transport systems represents one such area. An increased understanding of developmental processes that shape the kidney's unique physiology is vital in defining its response to injury and capacity for regeneration. The importance of this research in the context of regenerative medicine is highlighted by the growing incidence of chronic kidney disease [from approx. 10% (1988–1994) to 13.1% (1999–2004) in the U.S.A. alone] . Understanding the complexity of iminoglycinuria has been instrumental in dissecting the underlying mechanisms of mature renal physiology, confirming the idea that multiple systems exist for imino acids and glycine transport . However, our lack of knowledge of amino acid transport ontogeny has been highlighted by the unexplained phenomenon of developmental iminoglycinuria. It is this transient form of the disorder, long-attributed to delayed development of transport systems , that is a focus of the present study.
As we reported previously, the inherited form of iminoglycinuria is mainly caused by mutations in SLC36A2 and SLC6A20, with additional potentially modifying mutations occurring in SLC6A18 and SLC6A19 . In the present study, we have shown that Slc6a18 preferentially transports glycine and alanine in mice. However, these results are in variance with those of Singer et. al. , who suggested that Slc6a18 is rather a general neutral amino acid transporter and proposed that active transport requires co-expression with Ace2 but not collectrin. Although both studies agree that Slc6a18 accepts most neutral amino acids to some extent, with the exception of proline, the results of Singer et al. , regarding the interaction of Slc6a18 with auxiliary proteins, were surprising in view of the expression of Slc6a18 in the kidney where collectrin is prevalent. Furthermore, Slc6a19, which is closely related to Slc6a18, can interact with both collectrin and Ace2 for surface expression [16,18,20]. In contrast with the results reported by Singer et al. , we have found that Slc6a18 not only associates with Ace2, but also with collectrin. The auxiliary proteins used by the transporter can also explain the different substrate specificities found by the two studies. When co-expressed with Ace2, the substrate specificity of Slc6a18 was similar to that observed by Singer et al.  (results not shown). These important findings explain both the lack of Slc6a18 surface expression  and the very high fractional excretion of neutral amino acids (82%) in collectrin-deficient mice .
On the basis of the confirmation of the substrate specificity of Slc6a18 and the results of our previous study, demonstrating the involvement of SLC6A18, SLC6A19, SLC6A20 and SLC36A2 in the persistent form of iminoglycinuria , we have proposed in the present report that developmental iminoglycinuria is due to the reduced function of these same transporters and have now provided the first definitive evidence for this in mice. Furthermore, we have discounted the potential role of Slc36a1 in this disorder due to its expression in sub-apical compartments of proximal tubule cells. This expression pattern differs from the location of Slc36a1 on the apical membrane of intestinal enterocytes  and demonstrates that Slc36a1 does not contribute directly to the renal reabsorption of imino acids and glycine.
The involvement of these imino acid (Slc6a19, Slc6a20a and Slc36a2) and glycine (Slc6a18, Slc6a19 and Slc36a2) transporters in developmental iminoglycinuria was supported by their low protein and mRNA expression levels during the first week of life, which correlated with the detection of iminoglycinuria in all mice during this time. However, the Western blot analysis of Slc6a20a revealed a dominant higher molecular-mass isoform for this transporter at birth, which may represent an immature form that does not reach the apical membrane, as indicated by the lack of staining in the corresponding confocal immunofluorescence image (0 days). At 7 days post-birth, Slc36a2 and Slc6a18 were expressed on the apical membrane, along with Slc6a20a, and exhibited an increase in mRNA expression. However, despite this change in expression, hyperexcretion of imino acids and glycine in the urine of mice was still apparent and this may be explained by the persistently low level of protein expression during this time. Although a trend towards a positive correlation between mRNA and protein expression was observed for the majority of transporters throughout development, Western blotting results indicated a possible delay in the translation of mRNA into protein. Thus a complete set of mature transporters does not occur until 3 weeks after birth. However, in contrast with the SLC6 transporters, Slc36a2 exhibited a decrease in mRNA expression during the second week that was maintained at a consistent level until after the sixth week of life (Supplementary Figure S4 at http://www.BiochemJ.org/bj/428/bj4280397add.htm). Although the effects of this decrease did eventually manifest as a reduction in protein level by 6 weeks post-birth, the substantial delay between transcriptional changes and resultant cellular protein levels may be attributed to the slow turnover of mature Slc36a2.
It is likely that the observed changes in renal transporter expression and urinary phenotype of the mice are related to the combined effects of a reduced interaction with auxiliary proteins, the process of brush-border membrane development and the regulation of protein expression at a molecular level. Given that collectrin is required for optimal surface expression of Slc6a18, Slc6a19 and Slc6a20a , the delayed presence of these transporters on the apical membrane and low levels of protein expression may be attributed to the lack of expression and cytoplasmic distribution of collectrin during the first week of life. Furthermore, this may also explain the absence of Slc6a20a on the apical membrane at birth, despite the detection of a higher molecular-mass isoform in the Western blot analysis. Interestingly, Ace2 expression was observed to decrease following the increase in collectrin expression at day 14, indicating that Ace2 was displaced by collectrin as the kidney developed.
However, the additional influence of brush-border membrane development cannot be discounted. It is clear from morphological studies that nephrogenesis is not complete until approx. 34 weeks in humans [22,23] and 10 days post-birth in murine species [27,46]. Functional and structural maturation of the nephron and its luminal brush border continues thereafter as the proximal tubules increase in length [27,47]. Therefore, although we observed some apical expression of transporters in 1-week-old mice, their kidneys possessed fewer transporters overall compared with those of adult mice, which probably has an impact on the reabsorption capacity of the kidney. Furthermore, it is known from detailed morphological studies in the developing dog  and rabbit  that the S3 segment is the first proximal tubule segment to become morphologically mature and this is followed by lengthening and segmental differentiation of the proximal convoluted tubules (S1–S2). It is therefore likely that the delayed expression of transporters is partly governed by the segmental development of the nephron. However, apical expression of collectrin was not observed until the second week of life, despite the expression of other transporters for which we demonstrated overlapping distribution. This indicates that, although certain nephron segments may be morphologically distinct, regulation of protein expression at a molecular level also plays a role in transporter abundance and thus developmental iminoglycinuria.
Iminoglycinuria began to resolve during the second week of life, with a marked decrease in the excretion of proline and hydroxyproline observed for all mice examined at day 14. However, the decline in glycine excretion was more gradual and did not reach levels comparable with adult levels until the fourth week. These alterations in urinary phenotypes could be explained by the accompanying changes in transporter expression. During the second week of life, when a marked decrease in imino acid excretion was observed, there was a corresponding increase in the mRNA and protein expression levels of Slc6a19, Slc6a20a and collectrin. Furthermore, all three proteins were apically expressed. This indicates that the slow onset of these transporters and collectrin was responsible for both the hyperiminoaciduria observed during the first week and its resolution during the second week of life. Although there is some apparent redundancy with proline transporters, glycine reabsorption is mediated by a combination of Slc6a18, Slc6a19 and Slc36a2. Since immunofluorescence and Western blotting indicated that Slc36a2 expression was present early in life, it is likely that glycinuria resulted from the low expression of Slc6a18 and Slc6a19. This also represents the likely cause of the prolonged hyperexcretion of leucine, which is a major substrate of Slc6a19 and is also transported by Slc6a18.
These findings reinforce and extend several early studies into the reabsorption capacity of neonatal rat kidney. Similar to the present study in mice, a sharp decrease in imino acid excretion was seen in rats during the second week of life, whereas glycine hyperexcretion did not subside until the third week of life [32–34]. Furthermore, in cortical slices and isolated renal tubules from these developing rats, decreased substrate excretion was found to correlate with the increased activity of two independent high-affinity systems; one serving the imino acids that appeared during the second neonatal week and the other serving glycine which became active in the third week of life. This is consistent with our present mRNA and protein expression results, which demonstrated increased levels of the high-affinity imino acid transporter Slc6a20a during the second neonatal week. The high-affinity uptake of glycine detected by these rat studies may represent the increasing activity of Slc6a18. We observed an increase in both Slc6a18 mRNA and protein, with mRNA reaching levels approx. 6-fold higher than in adult kidney by the fourth week of life (see Supplementary Figure S5 at http://www.BiochemJ.org/bj/428/bj4280397add.htm). Furthermore, renal physiologists have identified the activity of a third low-affinity uptake system, common to imino acids and glycine, in both neonatal and mature rat kidney [29,32,34,49]. This system was found to transport glycine with higher affinity compared with other substrates, which is consistent with the known substrate preference of Slc36a2  as well as our expression data in the present study demonstrating apical expression of Slc36a2 from birth onwards.
In the present report, we have elucidated mechanisms of normal renal ontogeny by demonstrating that developmental iminoglycinuria in mice results from a transient uptake deficiency in the proximal nephron due to delayed expression of specific transport systems. In neonatal mice, we have demonstrated the involvement of Slc6a18, Slc6a19, Slc6a20a and Slc36a2, and the auxiliary proteins Ace2 and collectrin. Many transporters have been conserved throughout vertebrate evolution and a high degree of similarity exists between human and mouse orthologues, not only at the amino acid level, but also in terms of their functionality and distribution. For example, human and mouse SLC6A19/Slc6a19 [7,42], SLC6A20/Slc6a20a [14,51,52] and SLC36A2/Slc36a2  possess similar substrate specificities and transport mechanisms. In addition, the transporters we have identified in the present study as responsible for developmental iminoglycinuria in mice have also been shown by our group to be involved in the persistent inherited form of iminoglycinuria in humans . This not only suggests that the molecular mechanism of developmental iminoglycinuria in human infants also involves SLC6A18, SLC6A19, SLC6A20 and SLC36A2, but also emphasizes the relevance of these results to our understanding of normal human renal ontogeny.
Jessica Vanslambrouck wrote the manuscript and performed and analysed the qPCR, immunofluorescence and HPLC experiments; Angelika Bröer performed the Western blot analyses and oocyte experiments; Thuvaraka Thavyogarajah performed the electrophysiological studies; Jeff Holst assisted with the experimental design and edited the manuscript prior to submission; Charles Bailey assisted with the experimental design and design of immunopeptides and analysis; Stefan Bröer contributed to the design and supervision of Western blots and electrophysiological studies, and edited the manuscript prior to submission; and John Rasko conceived the project, designed the experiments, supervised and co-ordinated the study and edited the manuscript prior to submission.
This study was supported by funding from the National Health and Medical Research Council [grant number 402730], the Australian Research Council [grant number DP0559104], and the Cell and Gene Trust.
We are grateful to Chris Clarke (Waters Australia, NSW, Australia) for his advice regarding HPLC, Sarah Watson (Centenary Institute) for her assistance with the animal studies, Stephane Flamant (Centenary Institute) for his guidance regarding qPCR, and Adrian Smith (Centenary Institute) for his assistance with immunofluorescence confocal microscopy.
Abbreviations: Ace2, angiotensin-converting enzyme 2; BBMV, brush-border membrane vesicle; DAPI, 4′,6-diamidino-2-phenylindole; Gapdh, glyceraldehyde-3-phosphate dehydrogenase; GST, glutathione transferase; qPCR, quantitative real-time PCR; RT, reverse transcription; SLC6, solute carrier family 6; SLC36, solute carrier family 36; Tmem27, transmembrane protein 27 (collectrin)
- © The Authors Journal compilation © 2010 Biochemical Society