Cell Physiology


Transplacental transfer of taurine, a β-amino acid essential for fetal and neonatal development, constitutes the primary source of taurine for the fetus. Placental transport of taurine is compromised in pregnancies complicated by intrauterine growth restriction, resulting in a reduced concentration of taurine in cord plasma. This could impact on fetal cellular metabolism as taurine represents the most abundant intracellular amino acid in many fetal cell types. In the present study, we have used pure isolates of fetal platelets and T lymphocytes from cord blood of placentas, from normal, term pregnancies, as fetal cell types to examine the cellular uptake mechanisms for taurine by the system β transporter and have compared gene and protein expression for the taurine transporter protein (TAUT) in these two cell types. System β activity in fetal platelets was 15-fold higher compared with fetal T lymphocytes (P < 0.005), mirroring greater TAUT mRNA expression in platelets than T lymphocytes (P < 0.005). Cell-specific differences in TAUT protein moieties were detected with a doublet of 75 and 80 kDa in fetal platelets compared with 114 and 120 kDa in fetal T lymphocytes, with relatively higher expression in platelets. We conclude that greater system β activity in fetal platelets compared with T lymphocytes is the result of relatively greater TAUT mRNA and protein expression. This study represents the first characterization of amino acid transporters in fetal T lymphocytes.

  • cord blood cells
  • amino acid
  • taurine transporter protein
  • system β

taurine (2-aminoethanesulfonic acid) is the most abundant free amino acid in several tissues, including the placenta and the cellular components of blood (9, 21, 38, 41, 46). It is a β-amino acid with the sulfonic acid group attached to the β-carbon atom and is not incorporated into proteins (14). The physiological functions of taurine are diverse, including bile and xenobiotic conjugation, regulation of neuronal excitability, membrane protection, antioxidation, detoxification, and osmoregulation (22, 24, 31). The nutritional requirements for taurine are met both by dietary sources and biosynthesis from cysteine and methionine (53). However, in the human fetus and the neonate, taurine is an essential amino acid, as biosynthetic capacity is almost negligible (51) due to the lack of cysteine sulfonic acid decarboxylase, which catalyses the rate-limiting step in taurine synthesis (4). Transplacental transfer of taurine from maternal blood therefore constitutes the primary source of fetal taurine.

Taurine is transported across cell membranes, including those of the syncytiotrophoblast, the transporting epithelium of human placenta, by the system β-amino acid transport system which is specific for taurine and other β-amino acids such as β-alanine and hypotaurine (29, 32, 36). System β is a high-affinity, low-capacity transport system that is Na+- and Cl- dependent (10) with a 2:1:1 Na+:Cl:taurine stoichiometry (39, 48). System β activity has been demonstrated in many cell types and its catalytic activity is mediated by the taurine transporter protein (TAUT) (26, 43). TAUT is a 65–74 kDa protein with 12 transmembrane domains (TMDs), two to three N-glycosylation sites in the large extracellular loop between the third and fourth TMD, and a fourth N-glycosylation site on the loop between TMD XI and XII (26, 35, 54).

In the placenta, system β activity is present in both the maternal-facing microvillous membrane (MVM) and the fetal-facing basal membrane of the syncytiotrophoblast with greater activity in MVM (40). System β activity in the MVM is reduced by 33% in intrauterine growth restriction (IUGR) (12, 40) and the umbilical venous plasma taurine concentration in small-for-gestational age pregnancies is less than the plasma levels in appropriate-for-gestational age pregnancies (18). This reduction in system β activity in MVM of placentas of IUGR pregnancies is not underscored by a reduction in TAUT protein expression, suggesting discordance between system β activity and TAUT expression (46).

Taurine is the most abundant free amino acid in umbilical platelets (27) and fetal mononuclear cells (73–85% lymphocytes, 15–22% monocytes), constituting 44% of all free amino acids in mononuclear cells (21). The taurine concentration in cord platelets (1.84 μmol/ml) is 20 times higher than the concentration in cord blood plasma (0.11 μmol/ml) (27), with a similar trend in cord blood mononuclear cells (21). Thus cellular components of fetal blood accumulate taurine against a concentration gradient, suggesting the presence of an active transport system for taurine in these cell types.

System β activity has been demonstrated previously in adult platelets (1, 16), adult leucocytes (34), fetal platelets (27), and a mixed population of cord blood cells composed mainly of leucocytes and platelets (50). Recently, TAUT mRNA expression has been confirmed in adult platelets (42). Neither system β activity nor TAUT expression has been examined in other cellular components of fetal blood, such as lymphocytes or neutrophils, although the predisposition of the latter cell type to undergo activation favors the use of lymphocytes to examine basal transporter activity. In cord blood at birth, T lymphocytes (2.6–3.9 × 109/l) predominate over B lymphocytes (0.5–1.0 × 109/l) (17) conferring practical advantages with respect to their isolation and use as a fetal cell type.

Therefore, the goals of this study were the following: 1) to isolate pure T lymphocytes and platelets from cord blood for transport studies, and 2) to examine gene and protein expression for TAUT and the functional activity of system β in fetal T lymphocytes compared with fetal platelets isolated from cord blood of placentas from normal pregnancies at term. Our broader aim is to use these cells as models to examine whether transporter capacity is altered in fetal cells from IUGR pregnancies.


Isolation of fetal T lymphocytes and platelets.

All samples of cord blood or placental tissue were obtained with written informed consent as approved by the Local Ethics Committee. Placentas were obtained following vaginal deliveries or elective caesarean sections from term (37–42 wk) normal pregnancies. Cord blood from the chorionic vessels (20–60 ml) was obtained within 15 min of delivery of the placenta as 10 ml aliquots using a 21 G needle and placed in 25 ml sterile universal bottles with 400 units of sodium heparin without preservative (CP Pharmaceuticals, Wrexham, UK). The method used to isolate fetal T lymphocytes from cord blood was based on that of Elliot et al. (19) but with additional modifications. Each 10 ml cord blood was centrifuged at 500 g for 10 min at room temperature (Spinchron R Centrifuge, Beckman Coulter, High Wycombe, UK) and clear plasma was aspirated off. The cells were then made up to original volume and subsequently diluted 1:2 with Hanks' balanced salt solution (HBSS; GIBCO, Paisley, UK). This was then layered carefully onto 15 ml Ficoll-Hypaque (Lymphoprep, Axis-Shield, Upton, Huntingdon, UK) in 50 ml sterile tubes (Sarstedt, Leicester, UK) and centrifuged at 1,200 g for 20 min at room temperature without the brakes applied. The interface was removed by being carefully pipetted and washed with 20 ml of HBSS containing 2% newborn calf serum (GIBCO) first at 450 g for 10 min and subsequently at 220 g for 10 min.

The cells were then resuspended in 20 ml HBSS and layered onto 15 ml Lymphoprep and centrifuged at 1,200 g for 20 min without the brakes applied. The cells at the interface were washed with 20 ml PBS with 0.1% bovine serum albumin (BSA) at 450 g for 10 min and again at 220 g for 10 min. The cells were resuspended in 200 μl PBS containing 0.1% BSA. To this was added 40 μl fetal calf serum (GIBCO), 80 μl antibody mix from T lymphocyte negative isolation kit (Dynal Biotech, Bromborough, Wirral, UK), 73 μl CD42b antibody and 26 μl CD235a antibody (Dako Cytomation, Ely, UK) to abolish platelet and erythrocyte contamination, respectively. The cells were incubated at 4°C (for antibody binding) for 10 min and then washed with 2 ml of PBS containing 0.1% BSA at 500 g for 8 min. The cells were then resuspended in 1.8 ml PBS with 0.1% BSA. Immunomagnetic beads from the isolation kit were added to the cells, allowing 500 μl beads per 10 ml of cord blood, and incubated for 15 min at 20°C with gentle tilting and rotation (Tube rotator, Bibby Sterilin, Stone, Staffordshire, UK). The unwanted cells, namely erythrocytes, granulocytes, platelets, B lymphocytes, monocytes, and NK cells, formed rosettes with the beads. The rosettes were resuspended by being carefully pipetted 5–6 times, and 2 ml 0.1% BSA in PBS added to the incubate solution. The tube containing the cells was then placed in the magnetic particle concentrator, MPC-L for 2 min and the supernatant containing pure T lymphocytes was pipetted off, centrifuged at 220 g for 10 min, and the harvested T lymphocyte pellet suspended in appropriate buffer/medium.

To isolate fetal platelets, cord blood with sodium heparin was centrifuged at 150 g for 15 min to obtain platelet-rich plasma (1, 27). The upper half to two-thirds of platelet-rich plasma was aspirated with a fine pipette and washed twice with 10 ml 0.1% BSA in PBS at 1,500 g for 10 min. The platelet pellet was suspended in appropriate buffer/medium.

Cell purity markers.

The T lymphocyte pellets from each placenta were pooled and suspended in 150 μl 0.1% BSA in PBS. To three 10 μl aliquots, 500 μl 4.3% formalin in PBS was added, and stored at 4°C for 24 h. The cells were pelleted out and washed twice with 1 ml 0.1% BSA in PBS at 2,450 g for 5 min (Micro-centaur, Sanyo Gallenkamp, Loughborough, UK). The cells were resuspended in 90 μl 0.1% BSA in PBS and 5 μl mouse anti-human CD3/FITC (T cell marker) + CD19/RPE (B cell marker), 5 μl monoclonal mouse anti-human CD42b, platelet glycoprotein Ib/FITC (platelet marker), or 5 μl mouse IgG1/FITC negative control (all fluorochrome-labeled antibodies were supplied by Dako Cytomation) were added, respectively. The tubes were vortexed, pulsed down, and left in the dark at room temperature for 15 min. The tubes were then centrifuged at 2,450 g for 5 min and washed with 1 ml 0.1% BSA in PBS. The supernatant was discarded and the cells suspended in 500 μl 0.1% BSA in PBS and kept in the dark on ice until analysis with the flow cytometer. Similarly, platelets from each placenta were pooled and treated as above with a fourth aliquot labeled with 5 μl mouse anti-human CD235a/FITC (erythroid cell marker).

The flow cytometer (Coulter Epics Elite with Expo 32 MultiCOMP Software; Beckman Coulter, High Wycombe, UK) was calibrated for size and fluorescence before each analysis with fluorospheres (Dako Cytomation) and cells labeled with mouse IgG1/FITC negative control respectively. Samples were analyzed by forward scatter and side scatter (90°) of light. The samples were then analyzed by antibody fluorescence to determine T lymphocyte, B lymphocyte, platelet, and erythroid cell content. 5,000 cells (events) were acquired for each T lymphocyte isolate analysis and 10,000 cells for each platelet isolate analysis.

Cell viability.

T lymphocyte suspension (10 μl) was mixed with 90 μl 0.1% trypan blue in PBS. The suspension fluid (10 μl) was then mounted on the hemocytometer (Fisher Chemicals, Loughborough, UK) with coverslip. The slide was visualised under a ×20 objective with a microscope (Olympus CK2, Micro Instruments, Long Hanborough, UK). After 1 min, the stained and unstained cells were counted. Percentage viability was calculated as the number of unstained cells divided by the sum of the stained and unstained cells multiplied by 100.

System β activity.

After isolation, T lymphocytes and platelets were suspended in Tyrode buffer that contained 0.1% BSA [Tyrode buffer contained (in mM) 135 NaCl, 5 KCl, 1.8 CaCl2, 1 MgCl2·6H2O, 10 HEPES, and 5.6 glucose, pH 7.4]. Preliminary data had shown that lymphocytes required BSA to maintain viability. This has also previously been demonstrated for platelets (47). The cells were incubated in the buffer for 30 min before amino acid uptake to reduce intracellular concentration of amino acids and avoid trans-inhibition.

Cells (30 μl; 3–5 × 106 cells) were mixed with 20 μl Tyrode containing 0.1% BSA and 0.5 μM 3H-taurine (31.0 Ci/mmol; Amersham Biosciences, Little Chalfont, UK) to give an incubation volume of 50 μl containing 0.2 μM 3H-taurine. Uptakes were carried out at 37°C and were terminated by the addition of 1 ml ice-cold Tyrode buffer and vortexing. The 1 ml mixture was then filtered by the fast filtration technique through 0.45 μm HAWP filter (Millipore, Watford, UK) that was presoaked in Tyrode buffer. This was washed with 10 ml ice-cold Tyrode buffer to remove any extracellular isotope. To account for nonspecific binding of isotope to the filter, cells were replaced by 30 μl Tyrode containing 0.1% BSA representing no protein controls. Filters were placed in scintillation vials and dissolved with 2 ml 2-ethoxyethanol (BDH, Poole, UK). 12 ml Optiphase Hisafe II scintillant (Fisher Chemicals) was added and the radioactivity determined using a Packard Tri-Carb 2100TR liquid scintillation counter using standard window settings. Isotope standards (5 μl stock solution) were included along with counting backgrounds (2 ml ethoxyethanol and 12 ml scintillant). Uptake of 3H-taurine was expressed as fmol/106 cells.

To demonstrate sodium dependency of uptake, NaCl was replaced with choline chloride in Tyrode buffer. To show chloride dependency, NaCl, KCl, CaCl2, and MgCl2·6H2O were substituted with sodium gluconate, potassium gluconate, calcium gluconate, and MgSO4·6H2O, respectively. In this series of experiments, for T lymphocytes the incubating media contained 2 μM 3H-taurine. This was done because taurine uptake by lymphocytes at 0.2 μM was too low to delineate ion dependency of uptake.

For the competitive inhibition studies, 30 μl of cells were added to 10 μl 1 μM 3H-taurine (stock) and 10 μl 1 mM unlabeled taurine, hypotaurine, and β-alanine, respectively, to give an incubation mix 50 μl containing 0.2 μM 3H-taurine and 200 μM amino acid competitor. The competitive inhibition uptakes were undertaken at 37°C for 10 min.

TAUT mRNA expression.

Isolated fetal T lymphocytes and platelets were stored in RNAlater (Ambion) at −80°C before analysis. Fragments of placenta from term, normal placenta, were also harvested, washed in PBS, and stored in RNAlater, serving as a positive control/calibrator.

Total RNA was extracted, quantified, and cDNA synthesized as described previously (33). Briefly, RNA was extracted using Absolutely RNA (Stratagene, La Jolla, CA) with DNase I treatment to eliminate genomic DNA contamination. Total RNA was quantified in duplicate using Ribogreen assay (Molecular Probes, Invitrogen, Paisley, UK). A constant amount of total RNA (100 ng) from fetal T lymphocytes, platelets, or term placenta was used for reverse transcription (RT) with random primers (750 ng). Omission of RT enzyme or template (water-replacing RNA) served as negative controls.

TAUT and β-actin (housekeeping gene used to confirm amplification capacity) mRNA was quantified by real-time quantitative PCR using intercalation of SYBR Green as described previously (33). All reactions were run in duplicate and 5-carboxy-x-rhodamine was included as passive reference dye. The products generated were treated at 95°C for 1 min and then 55°C with increments of 0.5°C/30 s for 81 cycles to generate a melt curve. The sequences of the primers for TAUT were as follows: sense, 5′-CGTACCCCTGACCTACAACAAA-3′; antisense, 5′-CAGAGGCGGATGACGATGAC-3′. The sequences for the β-actin primers were sense, 5′-AGCCACCCCACTTCTCTCTAAG-3′; antisense, 5′-ACACGAAAGCAATGCTATCACCT-3′.

To quantify mRNA expression a standard curve was constructed from cDNA generated from human reference RNA (1 μg/ul; Stratagene) ranging from input amounts of 12 pg to 100 ng for β-actin and 1.95 pg to 100 ng for TAUT, over which there was a linear relationship between initial input cDNA amount and product cycle threshold values. cDNA input amount was interpolated from the standard curve using the sample cycle threshold value, as described previously (33). All sample values were normalized to term placenta as the internal calibrator.

TAUT protein expression.

Lysates of fetal T lymphocytes and platelets were obtained following homogenization in PBS containing 1% protease inhibitor cocktail (Sigma, Dorset, UK) on ice. Fetal T lymphocyte membranes were prepared by centrifugation of lymphocyte lysates at 100,000 g for 45 min at 4°C (Sorvall RC 28S, Thermo Electron, Bishop Stortford, UK). The pellet was suspended in 200 μl PBS containing 1% protease inhibitor cocktail and homogenized with a Dounce homogenizer on ice. JAr cell (a choriocarcinoma cell line) lysate and MVM isolates were included as positive controls for TAUT (13, 40). Protein concentration was determined using Bio-Rad dye reagent. Samples were loaded with a non-boil-reducing buffer in a 2:1 ratio as described previously (3). Sample viscosity was reduced by passage through a 21-G needle 10 times and polyacrylamide gel electrophoresis was performed in 30 μl wells using 3% stacking and 7% resolving gels and proteins electrotransferred to Hybond ECL nitrocellulose membrane as described previously (3).

Western blots were probed with a rabbit anti-taurine transporter antibody (1:400; Alpha Diagnostics International, San Antonio, TX) (37, 52, 55) with a blocking peptide (5–10 times excess) as a negative control. Goat anti-CD42b antibody (1:200; Santa Cruz, Insight Biotechnology, Wembley, UK) served as a positive control for platelets and mouse anti-β-actin antibody (1:1,000; Sigma) was used to confirm protein integrity. For TAUT and CD42b, antibodies were preabsorbed with blocking peptides overnight at 4°C. The membranes were blocked for 1 h at room temperature in Blotto (3% skimmed milk powder in TBS, pH 8, with 0.05% Tween) and were then probed with either TAUT (overnight at 4°C), CD42b, or β-actin (1 h at room temperature) antibodies in Blotto. Blots were then washed with 0.05% Tween in TBS. Species-specific horseradish peroxidase-conjugated IgG antibody (1:2,000 in Blotto; DAKO) was applied for 1 h at room temperature and visualization was performed using an ECL detection system (Amersham Biosciences).

Enzymatic deglycosylation of TAUT.

Fetal T lymphocyte, platelet and JAr cell lysates, and MVM isolates were denatured at 100°C for 10 min in 1× glycoprotein denaturing buffer (New England Biolabs, Hitchin, UK). The denatured proteins were then incubated at 37°C for 1 h in 1 × G7 reaction buffer, 1× Nonidet P-40, and 750–1,250 units of peptide N-glycosidase F (PNGase F; New England Biolabs) per 10 μg protein. For negative controls, PNGase F was replaced with equivalent volume of water. The proteins (at the same protein loading as above) were then separated by 7% SDS-PAGE and probed with anti-TAUT antibody. For TAUT deglycosylation in T lymphocytes and platelets, both treated and untreated samples were loaded onto 60 μl gel wells rather than the 30 μl wells used for the previous electrophoreses.

Statistical analysis.

Prism version 4.0 (GraphPad Software, San Diego, CA) was used to analyze the results. For the time course of uptake of 3H-taurine by system β in each cell type, the uptake at each time point is presented as means ± SE. Linear regression analysis was applied to determine linearity of uptake over 15 min. For ion dependency and competitive inhibition studies, data are presented as means ± SE. Friedman's test and Student's t-test, respectively, were used to determine significance of ion dependency and inhibition with respect to control values. Mann-Whitney test was used to determine the difference in relative gene expression for β-actin and TAUT in fetal T lymphocytes and platelets.


Purity of isolated fetal T lymphocytes and platelets.

The purity of the isolated fetal T lymphocytes and platelets was assessed by flow cytometry using CD3, CD19, CD42b, and CD235a as markers for T lymphocytes, B lymphocytes, platelets, and erythroid cells, respectively. The isolated T lymphocytes had light scatter properties characteristic of lymphocytes (Figs. 1 and 2) and constituted ∼82% of all flow cytometer events (gate A in Fig. 1). A smaller proportion, 16% of the flow cytometer events were cell fragments (gate B in Fig. 1), and of these ∼80% were CD3 positive. Thus the purity of intact cells in gate A was taken as a marker of T lymphocyte purity of the whole cell isolate (Fig. 2). In six cell isolates, 97 ± 2% (means ± SE) of the T lymphocyte isolates were positive for CD3 with no detectable contamination from B lymphocytes and platelets. The viability of the isolated cells was consistently >94%. The purity of the isolated platelets was 94 ± 2% (means ± SE, n = 3).

Fig. 1.

Flow cytometer dot plot of a fetal T lymphocyte isolate according to forward (FS) and side (SS) scatter of light. Cell population gated A had the scatter properties of lymphocytes, whereas cell population gated B did not have the scatter property of any of the cellular blood components, and were therefore presumed to be cellular fragments.

Fig. 2.

Flow cytometer dotplot analysis of cell population gated A (Fig. 1) according to CD3 positivity (FL1) and FS of light. 98.6% (F2) of this cell population were T lymphocytes (CD3 positive). This dot plot is representative of six isolates.

System β activity.

To determine whether system β activity was present in fetal T lymphocytes and platelets the time-dependent uptake of 0.2 μM 3H-taurine was measured in the presence of Na+. 3H-taurine uptake in T lymphocytes and platelets was linear up to 15 min as shown in Fig. 3, A and B. At 10 min, 3H-taurine uptake (means ± SE) in platelets (23.71 ± 6.00 fmol/106 cells; n = 5) was 15-fold greater (P < 0.005) than uptake in T lymphocytes (1.56 ± 0.18 fmol/106 cells; n = 6). In Na+ and Cl-free medium, 3H-taurine uptake in T lymphocytes and platelets at 10 min was significantly less than in control buffer (Table 1) demonstrating Na+ and Cl dependency of uptake. The number of T lymphocytes recovered from each isolation (∼12 × 106 cells) was not sufficient to allow uptakes to be performed both in the presence and absence of Na+ in a paired manner over the 15 min time course, with determination of the Na+-dependent component, so measurements over this time course were performed only in control Tyrode buffer with 0.1% BSA.

Fig. 3.

Time course of 0.2 μM 3H-taurine uptake in fetal T lymphocytes (A; n = 6, r2 = 0.995, P < 0.05) and fetal platelets (B; n = 5, r2 = 0.999, P < 0.05). Data are shown as means ± SE.

View this table:
Table 1.

Na+ and Cl dependence of 3H-taurine uptake in fetal T lymphocytes and platelets

As shown in Table 2, 3H-taurine uptake by T lymphocytes and platelets at 10 min was significantly (P < 0.05) inhibited by 1,000× excess unlabeled β-amino acids, including taurine, hypotaurine, and β-alanine, as would be expected for system β activity.

View this table:
Table 2.

Competitive inhibition of 3H-taurine uptake in fetal T lymphocytes and platelets

TAUT mRNA expression.

β-Actin and TAUT mRNA expression was detectable in both fetal platelets and T lymphocytes (Fig. 4). β-Actin mRNA expression was significantly higher (P < 0.005) in platelets compared with T lymphocytes (Fig. 4A). Similarly, TAUT mRNA expression in platelets, although more variable, was significantly greater (P < 0.005) than in T lymphocytes (Fig. 4B).

Fig. 4.

Relative gene expression for β-actin (A) and taurine transporter protein (TAUT; B) in fetal T lymphocytes and platelets. Data is shown as box and “whiskers” (box extends from the 25th to the 75th percentiles, whiskers show range with a horizontal line at the median, n = 6 for both). *P < 0.005 vs. platelets; Mann-Whitney test.

TAUT protein expression.

Probing of Western blots with anti-TAUT antibody revealed two immunoreactive species at 75 and 80 kDa in fetal platelets (Fig. 5A). The upper band co-migrated with the single TAUT protein band in JAr cells (at ∼80 kDa), whereas the lower band co-migrated with the TAUT protein band in MVM (at ∼75 kDa). There was also another TAUT band at ∼44 kDa in MVM. These immunoreactive bands were blocked by blocking peptide at 5× excess (Fig. 5B). TAUT signal was detected in four different platelet isolates (Figs. 5A and 8A), although signal was of variable intensity and was undetectable in PL2 at 50 μg protein loading (Fig. 5A). This variability in signal intensity was not mirrored when the same isolates were probed for CD42b as a specific platelet marker (Fig. 6). The intense immunoreactive signal for CD42b in each platelet isolate (at 133 kDa) confirmed their cellular identity as platelets (2) and suggested that protein integrity was preserved. This implies that the lack of TAUT signal in isolate PL2 and the doublet banding pattern observed for TAUT was not attributable to proteolytic fragmentation or degradation.

Fig. 5.

Western blot for TAUT in the presence of primary antibody alone (A; 1:400) or with blocking peptide (B; 1:80). Protein loadings were JAr cell lysate (10 μg), platelet lysates (PL1–4), and microvillous membrane (MVM; 50 μg). A: TAUT doublet at 75 and 80 kDa was present in PL1, 3 and 4 but no signal was detectable in PL2. B: TAUT signals at 75 and 80 kDa were abolished by excess blocking peptide; nonspecific bands at 222 and 119 kDa were visible with blocking peptide. Exposure time was 70 min in both A and B.

Fig. 6.

Western blot of four platelet lysates (PL1–4; 50 μg protein/lane) probed with CD42b antibody (1:200). A single CD42b band of similar intensity was visible in each platelet isolate. Exposure time was 1 min.

In fetal T lymphocytes, TAUT protein was detected as two major bands at 114 and 120 kDa, distinct from the TAUT species detected in JAr cells, fetal platelets, and MVM (Fig. 7A). A minor TAUT species at 78 kDa was also visible in T lymphocytes. All immuoreactive species were however, abolished in the presence of blocking peptide (Fig. 7B). Detection of signal in fetal T lymphocytes was variable between different isolations, and despite the relatively high protein loading of cell lysate (100 μg) used, was undetectable in some instances, even after long exposure times. To improve detection sensitivity, a membrane-enriched fraction was prepared from a total cell lysate of two fetal T lymphocyte isolates. A major TAUT species of 120 kDa was revealed in this fetal T lymphocyte membrane-enriched fraction (Fig. 8A), which was blocked by excess peptide (Fig. 8B).

Fig. 7.

Western blot for TAUT expression. Blot was probed in the presence of antibody alone (A; 1:400) or in the presence of excess blocking peptide (B; 1:50). Protein loadings were: JAr (10 μg), fetal platelet (PL) and fetal lymphocyte (FL) lysates (both 100 μg), and MVM (50 μg). A: major TAUT immunoreactive species were detected in all samples with a molecular mass of 80 kDa in JAr, a doublet of 78 and 80 kDa in PL, a doublet of 114 and 120 kDa in FL, and 44 and 78 kDa in MVM. B: signal was abolished by excess blocking peptide. Exposure time was 60 min in both A and B.

Fig. 8.

Western blot of membrane-enriched fraction of two fetal lymphocyte isolates (FLM1 and 2) probed with TAUT antibody alone (A; 1:400) or in the presence of excess blocking peptide (B; 1:40). Protein loadings were JAr (10 μg) and FLM1 and FLM2 (12 μg protein). A: intense TAUT signal was detected in both T lymphocyte membrane fractions at 120 kDa, with JAr signal (at 80 kDa) positive control. B: signal was abolished by excess blocking peptide. Exposure time was overnight in both A and B.

Treatment of the different cell lysates and MVM with PNGase F produced a shift in apparent molecular mass of the 75 kDa TAUT protein in MVM to 65 kDa but no change in the molecular mass of TAUT in JAr (Fig. 9A), lymphocyte or platelet lysates (Fig. 9B), although TAUT migration appeared altered under these conditions. All signals were abolished by blocking peptide (data not shown).

Fig. 9.

Western blot for JAr cell lysate and MVM (A); and (B) fetal lymphocyte (FL) and platelet (PL) lysates treated with (+) or without (−) peptide N-glycosidase F (PNGase F) and probed with anti-TAUT antibody (1:400). In JAr, TAUT immunoreactive signal at 82 kDa did not exhibit any mobility shift with PNGase F treatment while the TAUT band in MVM shifted from 77 to 65 kDa following deglycosylation. In lymphocyte and platelet lysates, there was no shift of the immunoreactive signals with PNGase F. The 55-kDa band present in untreated lymphocytes was not blocked by blocking peptide. Exposure time was 1 h.


This is the first report examining the activity of a transport protein in a pure population of fetal T lymphocytes. The absence of any previous study examining transporter activity in this cell type might be explained by the technical difficulties in isolating pure T lymphocytes from cord blood. Mononuclear cells have been isolated from peripheral blood by density gradient centrifugation over Ficoll-Hypaque (7, 8). While adult mononuclear cells isolated by this approach are contaminated with platelets (7, 20), mononuclear cells isolated from cord blood in this way in addition to platelets, are contaminated with mature erythrocytes and nucleated erythroid precursors (19). These have the potential to interfere with both functional and mRNA analyses (45). The importance of removing contaminating cell types is emphasized here; the collective data from the qPCR, Western blot analysis, and activity measurements demonstrates that fetal platelets differ greatly from fetal T lymphocytes with respect to TAUT gene and protein expression and system β activity, making it essential that the isolation technique is stringent enough to remove contaminating cell types that might act as confounders.

The procedure adopted here of two rounds of density-gradient centrifugation (to reduce erythroid cell contamination), followed by immunomagnetic bead depletion of each contaminating cell type with antibodies for erythroid cells (19), monocytes, platelets, B lymphocytes, NK cells, eosinophils, and basophils was successful in producing viable fetal T lymphocytes of high purity, as judged both by their light scattering properties (low forward and side scatter) and positive expression of CD3 antigen, a T lymphocyte cell surface marker (15), in ∼97% of isolated cells. We used negative immunodepletion of contaminating cell types, rather than positive immunoselection for T lymphocytes, as antibody binding to the cell surface might influence the physiology of the cell (23). This approach therefore confers the advantage of measuring transport activity in native, “untouched” cells.

We have previously reported system β activity in nonhomogenous fetal cord blood cell population composed of leucocytes, erythrocytes, and platelets (50). In the present study, we have been able to dissect further system β activity in two defined cell populations: fetal T lymphocytes and platelets isolated from the cord blood of normal, term babies at birth. We have demonstrated that fetal T lymphocytes and platelets accumulate 3H-taurine by a mechanism that is characteristic of system β. This is therefore the first data showing the existence of amino acid transport system β in a pure population of fetal T lymphocytes, although the activity in these cells is 15 times lower than that in fetal platelets.

Previous studies in fetal platelets have confirmed that taurine uptake is active and Na+-dependent (27), consistent with our observations. The lower system β activity observed in fetal T lymphocytes compared with fetal platelets is surprising in light of the relatively higher intracellular concentration of taurine found in the former (12 and 1.84 mM in fetal mononuclear cells and fetal platelets, respectively) (21, 27, 34). It should be noted that the value for the intracellular concentration of taurine in fetal mononuclear cells has been derived, based on the similar intracellular taurine content of fetal mononuclear cells and adult lymphocytes (26, 44) and on the assumption that both cell types have similar intracellular water content. The lower system β activity in fetal T lymphocytes is unlikely to be attributable to a faster dissipation rate of the Na+ gradient as the activity of system A, another Na+-dependent transporter, is higher in fetal T lymphocytes compared with platelets (25). This discordance between intracellular taurine concentration and system β activity has also been noted in primary neonatal rat heart cells, in which the taurine content in myocytes is higher than in nonmyocytes, while TAUT mRNA and protein expression is higher in the latter (52). In these cells, the transport capacity (Vmax) was higher in nonmyocytes with similar substrate affinity. Whether the difference in system β activity between fetal platelets and T lymphocytes reflects an altered substrate affinity or Vmax could not be addressed here. The number of cells isolated from each placenta imposed the constraint of being able to perform only four measurements at a cell density of 3–5 × 106 cells (confirmed to be optimal in preliminary studies; data not shown). However, the possibility of a higher Vmax in platelets is consistent with the greater TAUT gene and protein expression.

Although platelets are anucleate and lack nuclear DNA, they retain megakaryocyte-derived mRNA (5) and retain the ability for protein synthesis from cytoplasmic mRNA (6, 30). The purity of fetal platelets isolated and lack of contamination by leucocytes, demonstrated by flow cytometry, was also confirmed by measurement of RNA recovery which was 1,000 times lower than that extracted from a similar volume of T lymphocytes (data not shown). This, together with the differences observed in TAUT gene and protein expression and system β activity between fetal platelets and T lymphocytes, reinforces the purity and distinct characteristics of the two cell populations isolated. Our demonstration of TAUT gene expression in fetal platelets is consistent with previous observations confirming TAUT mRNA expression in adult platelets (11, 42).

The higher TAUT mRNA expression in fetal platelets compared with T lymphocytes was associated with more readily detectable TAUT protein expression in fetal platelets. TAUT expression was demonstrated in all but one of five platelet samples at 50 μg protein loading (though with variable signal intensity), but was only detectable in T lymphocyte lysates at a higher protein loading (100 μg). Membrane enrichment of T lymphocyte lysates improved sensitivity of TAUT detection, but gave a different banding pattern compared with that observed for T lymphocyte total cell lysate, with the 120-kDa TAUT species predominant.

Our demonstration that TAUT immunoreactive species have different molecular masses: MVM (44 and 75 kDa), fetal platelets (75 and 80 kDa), JAr cells (80 kDa), and fetal T lymphocytes (114 and 120 kDa) infers that the core protein might undergo posttranslational modification, such as glycosylation, in a cell-specific manner. There are four putative glycosylation sites in both human placental and thyroid TAUT (26). Previous studies using TAUT cDNA cloned from different cell types predicted molecular masses for TAUT protein of 65–74 kDa in the Madin-Darby canine kidney cell (54), rat brain (49), mouse brain (35), human thyroid (26), and human placenta (44). We have shown that in MVM, deglycosylation with PNGase F resulted in TAUT with the molecular mass of 65 kDa predicted for the core TAUT protein. This observation is consistent with studies in transfected baculovirus, where inhibition of TAUT glycosylation by tunicamycin resulted in TAUT of lower molecular weight (37).

The lack of a mobility shift in the molecular weight of TAUT in JAr, lymphocyte, and platelet lysates on treatment with PNGase F implies the absence of any N-glycan units in these TAUT moieties, or that the N-glycosidic bonds in these TAUT moieties are resistant to PNGase F. The ability of PNGase F to cleave the glycosidic bonds in TAUT in MVM and in RNase B, which was used as a positive control for N-deglycosylation (data not shown) confirmed that PNGase F was functionally active. A possible explanation for the failure of PNGase F-mediated cleavage could be the association of TAUT in JAr, T lymphocyte, and platelet lysates with chaperone proteins (such as calnexin, which is known to associate with TAUT) (37), thus rendering the TAUT moieties resistant to the denaturing process we have used before deglycosylation. This association with chaperone proteins might be lacking in plasma cell membranes, hence the deglycosylation of TAUT in MVM. Alternatively, these TAUT moieties in JAr, T lymphocytes, and platelets might not be the same TAUT core protein with differential glycosylation, but represent different isoforms of TAUT that share the same epitope to which the anti-TAUT antibody is raised. It should be noted that other TAUT species have been postulated (44), although the gene primers used by us generated one amplicon only. This is indicative of a single TAUT gene product.

The difference in TAUT molecular mass is not restricted to expression between different cell types, but is also apparent between plasma membranes of the same cell type as has been recently shown in syncytiotrophoblast of human placenta with TAUT detected as 70 kDa in MVM and 50 and 70 kDa in the basal membrane (46). Kang et al. (28) have also reported that TAUT is expressed as two bands of 70 and 66 kDa in immortalized rat brain capillary endothelial cells (TR-BBB13).

These differences in the molecular mass of TAUT protein might also be attributable to alternative splicing (56), expression of two different Na+-dependent taurine transporters (44), and differential phosphorylation of the peptide, as both human placental and thyroid TAUT have several sites for phosphorylation, which might be involved in regulation of activity (26, 44). In addition, it has been suggested that the molecular mass of TAUT protein might be dependent on the subcellular localization of the transporter (55). In mouse NIH3T3 fibroblasts, whole cell lysate and cytosolic TAUT protein was found to have a molecular mass of 67 kDa, while the transporter protein localized to the nucleus and cell membrane has a mass of 90 kDa (55). The TAUT doublet present in fetal platelets and T lymphocytes, and the differences in their molecular masses might therefore reflect cellular compartmentalization with differential posttranslational modification of TAUT in the different compartments within these cell types. In support of this notion, only a single TAUT 120 kDa band was observed in a membrane-enriched fraction of T lymphocyte membranes, compared with a doublet of 114 and 120 kDa in T lymphocyte cell lysate, suggesting the 114 kDa TAUT species may be localized to the cytosol.

In summary, we have demonstrated system β activity in fetal T lymphocytes and platelets isolated from the cord blood of term normal pregnancies. This activity is mediated by TAUT which is expressed at both mRNA and protein levels in both cell types. TAUT is present as 75- and 80-kDa doublet in platelets and 114 and 120 kDa in lymphocytes with the 120-kDa protein present in membrane-enriched fractions of lymphocytes. The different TAUT species might represent differential glycosylation of the core TAUT protein or different isoforms of TAUT. Platelets have a higher mRNA and protein expression of TAUT than T lymphocytes and greater system β activity.


This study was supported by the Wellcome Trust.


We thank the midwives and medical staff at St. Mary's Hospital, Manchester, for help in obtaining placentas. We also thank Prof. H. Zola (Child Health Research Institute, University of Adelaide, Australia) for advice concerning the fetal T lymphocyte isolation procedure, and Dr. H. Lacey for guidance with the real-time qPCR studies.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract