Original Article

European Journal of Clinical Nutrition (2006) 60, 1–8. doi:10.1038/sj.ejcn.1602260; published online 24 August 2005

Influence of tea drinking on manganese intake, manganese status and leucocyte expression of MnSOD and cytosolic aminopeptidase P

Guarantor: JJ Powell.

Contributors: JJP is a nutritional scientist and he devised the overall study with help from KLG (nutritionist). KD is a molecular biologist and he devised and undertook the preparation of RNA/cDNA and the PCR analysis with help from S-JH (PhD student) and JJP. MN is a nutritional epidemiologist and he devised and advised on the FFQ including its administration and analysis, which were undertaken by S-JH with help on the administration from KLG. S-JH and SC (analytical chemist) devised the methodology for and undertook all Mn analyses. S-JH and KLG recruited and venesected the volunteers. S-JH prepared the blood, plasma and white cells. All authors contributed to data analysis, which was coordinated by JJP. All authors contributed to the writing and the editing of the manuscript, which was coordinated by SJ-H and JJP.

S-J Hope1, K Daniel2, K L Gleason3, S Comber1, M Nelson3 and J J Powell3,4

  1. 1WRc-NSF Ltd, Henley Road, Medmenham, Marlow, Buckinghamshire, UK
  2. 2Kings College London, Department of Immunology, The Rayne Institute, Denmark Hill, London, UK
  3. 3Kings College London, Department of Nutrition and Dietetics, 150 Stamford Street, London, UK
  4. 4MRC Human Nutrition Research, Elsie Widdowson Laboratory, Fulbourn Road, Cambridge, UK

Correspondence: Professor JJ Powell, MRC Human Nutrition Research, Elsie Widdowson Laboratory, Fulbourn Road, Cambridge CB1 9NL, UK. E-mail: jonathan.powell@mrc-hnr.cam.ac.uk

Received 5 October 2004; Revised 23 May 2005; Accepted 16 June 2005; Published online 24 August 2005.





Since black tea contains high levels of manganese (Mn), we investigated the relationship between dietary Mn intake, circulating Mn levels and leucocyte expression of two Mn-dependent enzymes in tea drinkers and non-tea drinkers.



We assessed Mn intakes (food frequency questionnaire), fasting whole blood and plasma Mn levels, and quantitative expression of peripheral blood mononuclear cell Mn-dependent superoxide dismutase (MnSOD) and cytosolic aminopeptidase-P (cAP-P).

Setting and subjects:


In total, 24 tea drinkers (greater than or equal to1 l black tea/day) and 28 non-tea drinkers were recruited from the staff and students of King's College London by circular email.



Dietary Mn intakes (mean (range)) were significantly lower (P<0.0001) in non tea drinkers (3.2 mg/day (0.5–6.5)) than tea drinkers (5.5 mg/day (2–12) or 10 mg/day (5–20) depending upon the value used for Mn levels of black tea). Whole blood, plasma Mn levels and expression of MnSOD and cAP-P did not differ between the groups. In a continuous analysis, whole blood Mn levels and expression of MnSOD correlated inversely but no other parameters associated with each other.



Tea drinking is a major source of dietary Mn and intakes commonly exceed proposed adequate intake values of 1.8–2.3 mg Mn/day and, on occasion, exceed upper limits of 10–11 mg/day. Dietary Mn intake has little influence on markers of Mn status or expression of Mn-dependent enzymes. Fasting whole blood Mn levels and leucocyte expression of MnSOD could, together, be further investigated as markers of Mn status.



S-JH was supported through the EPSRC PTP scheme. Running costs were from the UK Technical Tea Trade Association.


manganese, intake, status, tea, superoxide dismutase, cytosolic aminopeptidase P



Tea is the second most popular beverage in the world (International Tea Committee, 1995; Wrobel et al., 2000) and about 50–60 billion cups of tea are consumed per annum in each of the USA (The American Tea Society, 2002) and the UK (The Tea Council, 2003a, 2003b). The latter has a much higher per capita intake at about 3.5–4.0 cups/person/day (Ulteamate, 2003) although the consumption of tea is increasing in the USA (International Tea Committee, 1995). Leaf tea, which is used to make the infusion, is produced from fermentation and/or drying of the tea plant leaves. Unfermented tea is referred to as green tea, partially fermented is termed oolong tea while fully fermented is termed black tea (The Tea Council, 2003a, 2003b). Black tea is by far the commonest form of tea that is drunk in the Western world and the infusion is rich in certain nutrients, especially minerals (Powell et al., 1998).

Tea drinking is a potentially major source of dietary manganese (Mn) as 1 l (four standard mugs) of black tea is estimated to contain 1.8–5.2 mg Mn (Gillies and Birkbeck, 1983; Fraile and Flynn, 1991; Matsushima et al., 1993; Powell et al., 1998; Matsuura et al., 2001).

Mn is an essential element for a number of key enzymes including liver pyruvate carboxylase, arginase and, most notably, mitochondrial or Mn-dependent superoxide dismutase (MnSOD) (Greger, 1998). MnSOD is expressed in all cells of the body and plays a fundamental role in the destruction of mitochondrial superoxide anions, which are by-products of normal cellular respiration. The essential nature of MnSOD is evidenced from the finding that knock out mice, homozygous for the MnSOD gene deletion (MnSOD (-/-)), are unable to survive beyond a few weeks of age (Macmillan-Crow and Cruthirds, 2001). Even mice that are heterozygous for the mutation (i.e. MnSOD (-/+) mice) are prone to mitochondrial oxidative damage and dysfunction (Macmillan-Crow and Cruthirds, 2001). In contrast, overexpression of MnSOD, in hepatic mitochondria of rats, affords remarkable protection against high-dose ethanol probably by preventing the formation of free radical adducts (Wheeler et al., 2001).

In rats, increased Mn intake increases MnSOD in certain tissues (Thompson et al., 1992; Malecki et al., 1994), while, in human subjects, Mn supplementation increases the activity of MnSOD in circulating leucocytes (Davis and Greger, 1992). It is not known, however, how the intake of dietary Mn relates to cellular MnSOD levels in free living subjects although, in combination with circulating levels of Mn, leucocyte MnSOD activity has been suggested as a biomarker of Mn status (Greger, 1998, 1999). Both plasma (Greger, 1998, 1999) and whole blood (Clegg et al., 1986) have been proposed as appropriate compartments for the assessment of circulating Mn levels.

In addition, the recently described mammalian cytosolic aminopeptidase P (cAP-P) is a Mn-dependent enzyme that is expressed in leucocytes (Cottrell et al., 2000). This enzyme appears to be involved in a discrete pathway, namely the hydrolysis of bradykinin and substance P (Cottrell et al., 2000; Kulkarni and Deobagkar, 2002), and thus could be less prone to activation by multiple cellular stimuli than MnSOD (Greger, 1998; Macmillan-Crow and Cruthirds, 2001).

In spite of the relatively broad array of Mn-dependent enzymes in human physiological function, there is no Recommended Dietary Allowance in the USA or Recommended Nutrient Intake in the UK for Mn. For adults, in the UK, safe intakes for Mn are believed to be above 1.4 mg/day while the EU Scientific Committee on Food has only set a wide, acceptable range of 1–10 mg Mn/day for adults (SCF, 2000). In the USA, Adequate Intakes (AI) of 1.8 mg (F) and 2.3 mg (M)/day have been suggested for adults (IOM, 2001).

The objective of this study was to determine whether potential markers of Mn status differ between non-tea drinkers and tea drinkers, as the latter were expected to have higher Mn intakes. Hence, comparisons were made between the two groups for daily dietary intakes of Mn, levels of Mn in whole blood and plasma, and leucocyte expression of two Mn-dependant enzymes, namely MnSOD and cAP-P. For the enzyme assays, quantitative mRNA expression was preferred to measures of activity, as this is a simpler technique that, potentially, could then be more widely applied in subsequent studies.


Materials and methods


The project was approved by King's College ethics research committee and all subjects gave written consent. Volunteers, not taking nutritional supplements and consisting of staff and students from King's College London, were recruited by circular e-mail that asked for tea drinkers and non-tea drinkers to attend the Nutrition Department to provide a venous blood sample and to complete a food frequency questionnaire (FFQ). A total of 52 healthy volunteers were recruited, 24 black-tea drinkers (age range 22–62 years) and 28 non-tea drinkers (age range 21–63 years) (Table 1) and neither group consumed green tea or fruit/herbal teas. Tea drinkers were defined as those who consumed four or more mugs (i.e. greater than or equal to1 l) of black tea daily. Non-tea drinkers were those who did not consume tea at all.


Since no validated FFQ has been developed to determine dietary intakes of Mn, we used a general FFQ that has been designed to capture the variation in dietary intakes of the UK population including most vitamins and minerals (Nelson et al., 1997). The FFQ consisted of a 20-page paper questionnaire and asked subjects about their usual consumption of 111 food items ranging from bread and cereal to meat, vegetables and drinks. In total, 10 frequency categories for each food item were used; 1–7 days/week, fortnightly, monthly or never. Volunteers were asked to quantify the amounts of each food consumed in terms of household measures (Nelson et al., 1997). A manuscript on the validation of the FFQ is in preparation.

The weight of each food was calculated using the IDA (Integrated Dietary Analysis) software package (IDA, 1997) and by using a compilation of typical weights and portion sizes of foods eaten in the UK (MAFF, 2001). Dietary intakes of Mn were calculated using the McCance and Widdowson food composition tables (Holland et al., 1991). Daily Mn intakes were determined by multiplying the consumption frequency of each food by the nutrient content of the assigned portion size and by summing these values for all foods consumed by each subject (Tsubono et al., 2001).

Collection of blood samples

Overnight fasting venous blood samples were collected from 52 volunteers (n=28 non-tea drinkers and n=24 tea drinkers) in sterile 10 ml vacutainers containing lithium heparin (Becton Dickinson Vacutainer Systems, Plymouth, UK) as an anticoagulant. In total, 20 ml of blood were collected. A volume of 4 ml was taken for whole blood Mn analysis and 6 ml for plasma Mn analysis following centrifugation and transfer of plasma to a clean polypropylene container. Plasma samples were without visible haemolysis and both plasma and blood samples were stored at -20°C until analysis.

The remaining 10 ml of blood was collected for separation of peripheral blood mononuclear cells (PBMNC) and estimation of mRNA expression of two Mn-dependent enzymes, namely MnSOD and cytosolic aminopeptidase-P (cAP-P). Gene transcription of hypoxanthine-guanine phosphoribosyl transferase (HPRT) was also assessed as a reference (housekeeping) gene.

Manganese content of black tea

Black tea infusions were prepared using four popular UK brands (Yorkshire, PG, Tetley and Safeway). For each variety, 250 ml of boiling tap water were added to one tea bag in a clean mug for 1 min. The bag was then removed and the infusion allowed to cool before analysis for Mn content using a Thermo Jarrell Ash VIDEO 12E flame-atomic absorption spectrometer. For each black tea brand, a total of 11 tea bags were analysed in duplicate. The reference material used was NIES No7 tea leaves with a certified value of 700plusminus25 mug/g Mn and the mean value from analysis of this reference material was 696plusminus25 mug/g Mn (Hope, 2005).

Mugs were washed in nitric acid and rinsed extensively with deionized water prior to preparing the tea beverage. The Mn content of the tap water used to prepare the tea was substantially lower than that of the tea itself (Thames Water; <5 mug/l Mn) and its contribution to the overall Mn content of the tea beverage was, therefore, minimal (i.e. <1%).

Whole blood and plasma analysis

A range of branded needles, syringes and vacutainers was tested for contamination of leachable Mn. A 0.1% HNO3 solution (BDH, Poole, England) was drawn into syringes, using the needles, to maximum volume. Vacutainer tubes were also filled to maximum volume and all were left overnight and then the solutions analysed by graphite furnace-atomic absorption spectrophotometry (GF-AAS). Based upon these data, levels of contamination did not exceed 0.26plusminus0.2 ng Mn/ml in any sample. Standard vacutainers and 20-gauge needles yielded undetectable levels of contaminant Mn and were therefore used in the volunteer studies.

Working standards were prepared from atomic absorption spectrophotometry (AAS) standard solutions (Spectrosol, BDH, Poole, England) while all other reagents were of analytical grade. Plasma and whole blood samples were thawed at room temperature and diluted (1:3) using 0.1% Triton X-100 and 0.2% HNO3 in deionized water. A Pd(NO3)2 matrix modifier solution containing 3000 mg/l Pd was prepared (Schlemmer and Welz, 1986). Mn determinations of whole blood and plasma were performed using a Perkin Elmer 4100ZL GF-AAS with Zeeman effect background correction using an adapted method of Luna and Campos (1999). Automated injections were made using a Perkin Elmer AS-70 that delivered 10 mul of matrix modifier to 20 mul of sample. Pyrolytically coated tubes with platforms were used for all determinations. The accuracy and validity of the analytical data were established through the use of standard addition calibration, triplicate analysis and analysis of certified reference materials. Standard additions of 0, 2.5 and 5 mug/l Mn were used for plasma analysis and additions of 0, 5 and 10 mug/l Mn were used for whole bloods. Standard additions were made using a diluted stock solution from a standard Mn solution (Spectrosol, BDH, Poole, England). Certified reference materials were Seronorm Trace Elements, Serum (certified value 10.6 ng/ml Mn with an analytical range of 10.2–11.2 ng/ml) and Seronorm Trace Elements, Whole Blood, Level 2 (certified value 13.4 ng/ml Mn with an analytical range of 12.8–15.1 ng/ml) (both SERO AS, Norway).

Determination of MnSOD and cAP-P expression

RNA was isolated from total PBMNC using Trizol reagent (Invitrogen Technologies, USA). Total RNA was measured using spectrophotometry (A260 nm) and, for each sample, 5 mug total RNA was used to make cDNA by reverse transcription oligo-dT priming using the MMLV-RT enzyme (Invitrogen Technologies) according to the manufacturer's recommendations. Total cDNA was diluted 1:3 and used for two-step real-time-polymerase chain reaction (PCR). Real-time PCR was carried out in a total volume of 20 mul using Qiagen's SYBR-green PCR mastermix and 2 mul of the diluted cDNA. The primers, designed in-house and analysed for compatibility by means of Blast searches, were as follows: AACGTCACCGAGGAGAAGTACC as the sense primer for human MnSOD and CCTTGGACACCAACAGATGC as the antisense primer. The size of the amplified product was confirmed as 245 bp, and annealed at 55°C. For human cAP-P, the sense primer was AGTGACAAGGCCAGCTATGC and TGAGATCTCTGTCACACCACC for the antisense primer. The size of the amplified product was confirmed as 210 bp, and annealed at 55°C. The reference (housekeeping) gene was human HPRT with a sense primer of TTGTAGCCCTCTGTGTGCTCAAG and an antisense primer of GCCTGACCAAGGAAAGCAAAGTC. The size of the amplified product was confirmed as 270 bp, and annealed at 60°C. Amplification and instrument settings were according to the manufacturer's recommendation (Qiagen SYBR Green PCR mix), using the Light Cycler real-time PCR instrument (Roche). In brief, a 15 min denaturation was followed by 30 cycles of 20 s denaturation at 94°C, 20 s annealing at 55–60°C, 30 s extension at 72°C, followed by a melting curve analysis. For analysis of real-time PCR data, identical serial dilutions (in 10-fold steps from 100 to 0.01 pg per reaction) of a plasmid (Litmus 38, New England Biolabs), containing the relevant PCR product cloned in the EcoRV site, were used in every single run as a standard curve. The calculated values of MnSOD and cAP-P were normalized against the calculated values of HPRT expression for every sample and this ratio is expressed as a relative value (rv) with the lowest being set as 1.0.

Statistical methods

Categorical variables (i.e. results from tea drinkers and non-tea drinkers) were compared for Mn intake, fasting plasma and whole blood Mn levels and expression of leucocyte MnSOD and cAP-P. A priori hypotheses were that Mn intake, circulating Mn levels (plasma and whole blood) and expression of leucocyte Mn-dependent enzymes would be greater in tea drinkers than non-tea drinkers. Data were compared by unpaired t-test if normally distributed and Mann–Whitney test if non-normally distributed. Owing to the interest and uncertainty in potential biomarkers of Mn status in humans, we also undertook correlations of the data as continuous variables (i.e. not stratified for tea drinking).



The meanplusminuss.d. Mn level of the certified whole blood reference material was 13.5plusminus1.3 ng/ml (expected value 13.4 ng/ml), and the mean value of the serum reference material was 10.8plusminus0.5 ng/ml (expected value 10.6 ng/ml).

Mn levels of black tea infusions were 0.51plusminus0.11 mg/100 g (Meanplusminuss.d.) in agreement with previous findings (Powell et al., 1998) but higher than the 0.14 mg/100 g used in the McCance and Widdowson UK food composition database (Holland et al., 1991). However, Mn intake (mean (range)) was significantly greater in tea drinkers than non-tea drinkers using either the value of 0.51 mg/100 g (10 mg/day (5–20) versus 3.2 mg/day (0.5–6.5), respectively; P<0.0001 by t-test) or 0.14 mg/100 g (5.5 mg/day (2–12) versus 3.2 mg/day (0.5–6.5), respectively; P<0.0001 by t-test) (Figure 1).

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Total daily dietary manganese (Mn) intakes for tea drinkers and non-tea drinkers assuming, as found by analysis, a level of 0.51 mg Mn/100 ml black tea infusion (main graph) or, as quoted in the UK food composition database, a level of 0.14 mg Mn/100 ml black tea infusion (inset). In both cases, P<0.0001 for Mn intakes of tea drinkers versus non-tea drinkers.

Full figure and legend (42K)

Mean fasting plasma Mn levels were similar (Mann–Whitney test) in both groups (P=0.94) varying from 0.1 to 4.4 ng/ml (Figure 2). Whole blood Mn levels (Mann–Whitney test) and expression of cAP-P (t-test) and MnSOD (Mann–Whitney test) did not differ significantly in tea drinkers compared with non-tea drinkers (P=0.11, 0.15 and 0.4, respectively; Figures 3, 4 and 5). Overall, fasting whole blood Mn levels (Figure 3) were about six-fold higher than fasting plasma Mn levels (Figure 2). The expression of cAP-P was normally distributed (Figure 4) and fell within a relatively narrow band (1.0–8.2 units for all subjects) in contrast to the expression of MnSOD (Figure 5), which varied by more than two orders of magnitude between subjects.

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Distribution of plasma manganese (Mn) levels (ng/ml) in non-tea drinkers compared to tea drinkers (P=0.94). – indicates the mean.

Full figure and legend (31K)

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Distribution of whole blood manganese (Mn) levels (ng/ml) in non-tea drinkers compared to tea drinkers (P=0.11). – indicates the mean.

Full figure and legend (30K)

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Distribution of PBMNC cAP-P expression in relative units (rvcAP-P), from real-time PCR analysis, in non-tea drinkers compared to tea drinkers (P=0.15). – indicates the mean.

Full figure and legend (30K)

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Distribution of PBMNC MnSOD expression in relative units (rvMnSOD) from real-time PCR analysis in non-tea drinkers and tea drinkers.

Full figure and legend (30K)

Relationships between the expression of Mn-dependent enzymes and Mn intake or circulating Mn levels were investigated by correlation. Following exclusion of an outlying point at x=26.8 and y=69.7 (see Figure 6), the expression of MnSOD correlated, inversely, with whole blood Mn (r=-0.36, P=0.02; simple linear regression) although the association appeared to follow a nonlinear relationship (Figure 6). The best-fit curvilinear relationship was given by the expression: MnSOD=-2.8+141.9/whole blood Mn, with r=0.41 and P=0.002.

Figure 6.
Figure 6 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Associations between MnSOD expression and blood Mn levels (top) and between cAP-P expression and Mn intakes (bottom) in all subjects. Squares=tea drinkers; circles=non-tea drinkers. See Results section for statistics.

Full figure and legend (48K)

The linear correlation between expression of cAP-P and Mn intake did not achieve significance (r=0.2, P=0.16; Figure 6). There was no association (Pgreater than or equal to0.5) between the remaining combination of variables by correlation.



There is no Recommended Dietary Allowance in the USA or Recommended Nutrient Intake in the UK for Mn. For adults, in the UK, safe intakes for Mn are believed to be above 1.4 mg/day while the EU Scientific Committee on Food has only set a wide, acceptable range of 1–10 mg Mn/day for adults (SCF, 2000). In the USA, AI of 1.8 mg (females) and 2.3 mg (males) per day have been suggested for adults (IOM, 2001). In all cases, however, there has been considerable uncertainty in setting these Dietary Reference Values/Intakes and in specifying with any confidence the percentage of individuals covered by the recommended intake (IOM, 2001).

Reasons for this uncertainty are multiple. First, although dietary Mn intake data are commonly available, such as from the UK National Diet and Nutrition Surveys (NDNS) (Finch et al., 1998; Gregory et al., 2000; Henderson et al., 2003b), the food composition tables that serve these surveys do not always have comprehensive and contemporary data for the Mn contents of foods. Secondly, there are too few detailed studies of dietary Mn turnover and metabolism (Freeland-Graves et al., 1988; Greger et al., 1990; Davis and Greger, 1992). Thirdly, there is a perception that human Mn deficiency has not been observed in the population (IOM, 2001) and yet, as noted above, not only is true deficiency rare for most nutrients in the UK, it is not even known how long-term, suboptimal Mn intake would manifest itself. Finally, there are currently no useful biological indicators of dietary exposure or status for Mn (IOM, 2001).

In the present study, the estimates of Mn intake were based on an FFQ that asked about consumption over the previous year. The precision of this estimate in relation to true intakes is difficult to establish, given that there is no independent reference measure for Mn against which to compare estimated intake (e.g. a robust biological marker) and the general difficulties of dietary assessment and its interpretation (Nelson et al., 2004). The recent adult NDNS (Henderson et al., 2003a) reported mean (s.d.) Mn intake from food sources in men was 3.32 mg/day (1.42) and in women 2.69 (1.10), based on 7-day-weighed inventories of diet and using the McCance and Widdowson figure of 0.14 mg/100 ml infusion for black tea. The present estimates of Mn intake (3.2 mg/day in non-tea drinkers and 5.5 mg/day in tea drinkers) are of the same order of magnitude as the NDNS values. The slightly higher values from the present FFQ do not necessarily represent a bias towards over-reporting, as there was substantial under-reporting (of energy) in the NDNS (Henderson et al., 2003b), while tea drinkers in the current study consumed greater than or equal to4 cups tea/day and hence Mn intakes in this group are expectedly higher.

In this work, for non-tea drinkers, the mean values of Mn intake were 3.2 mg/day (range 0.5–6.5 mg/day) and, therefore, consistent with acceptable dietary intakes (EU) or slightly higher than the AI (USA). For moderate to heavy tea drinkers, defined in this study as those consuming at least 1 l (four mugs) black tea infusion per day, the values were significantly higher and dependent upon the concentration used for the Mn content of tea infusions. Previously, figures around 0.5 mg/100 g have been reported for the Mn content of black tea infusions although those as low as 0.1 mg/100 g have also been suggested (Gillies and Birkbeck, 1983; Matsushima et al., 1993). Clearly the type of tea, including its geographical origin, degree of fermentation prior to drying and leaf size, as well as type and temperature of water, length of brewing time and strength of tea brew will all impact upon the concentration of Mn and other nutrients present in the infusion. The McCance and Widdowson value of 0.14 mg/100 g (Holland et al., 1991) would appear low but would mean that, in this study, tea drinkers ingested an average of 5.5 mg Mn/day (range 2–12 mg/day). Our limited studies with typical commercial tea bags, steeped in boiling water in a mug for 1 min, as most typically occurs in the UK, yielded average values of 0.51 mg/100 g and hence this value was used in further analysis of the data. In this case, 'tea drinkers' ingested a mean of 10 mg Mn/day (range 5–20 mg/day), greatly exceeding the AI (USA). In fact, approximately 40% of moderate to heavy tea drinkers exceeded the acceptable dietary intake (EU) of 10 mg and the Tolerable Upper Intake Level (UL) (USA) of 11 mg Mn/day (IOM, 2001). Tea polyphenols have relatively weak affinity for most divalent metal ions under gastrointestinal conditions, including Mn2+ (Powell et al., 1998), and in vitro studies, using human gastric juice (Powell et al., 1998), as well as feeding studies in animals (Fraile and Flynn, 1991), suggest that Mn is potentially well absorbed from tea. Taken together, these data suggest that the AI and UL of 1.8–2.3 and 11 mg Mn/day respectively, and the upper end of the SCF's acceptable dietary intake (10 mg Mn/day), are relatively conservative values, especially for tea drinkers.

Much of this study considered the effect of habitual high Mn intakes from tea drinking on circulating Mn levels and leucocyte expression of Mn-dependent enzymes. Neither outcome was affected significantly by tea drinking and there are several possible interpretations of these findings. First, the additional Mn intake from tea drinking may not contribute to the dietary fraction of bioavailable Mn. However, as noted above, limited in vitro and in vivo studies suggest otherwise, although Mn absorption studies from tea drinking in human subjects are clearly required to quantify this. Secondly, efficient Mn homoestasis, which occurs mostly at the level of the gut (liver/gastrointestinal tract) (Clegg et al., 1986), may maintain a similar Mn status in individuals habitually ingesting 0.8–20 mg Mn/day (i.e. the range of Mn intakes in this study) (Finley et al., 2003). Indeed, most studies that have observed changes in plasma or whole blood Mn levels or changes in cellular MnSOD have been concerned with conditions of Mn depletion or supplementation. It is possible, therefore, that Mn supplementation (or markedly increased dietary intake) only impacts upon the parameters studied here in previously depleted individuals. Again, further work should address this issue. Thirdly, at least for Mn-dependent enzyme expression, it is possible that mRNA does not reflect the enzyme concentration, activity or its utilization of Mn (Macmillan-Crow and Cruthirds, 2001). Previous work with cellular uptake of metal ions and investigation of metalloenzyme mRNA expression shows that the linear response of gene expression deviates markedly with very high or very low levels of the enzyme-specific metal ion (Fanzo et al., 2001). Nonetheless, this work is based upon findings with habitual dietary intakes and both blood and plasma Mn levels were normally distributed.

It was interesting to note that the expression of leucocyte MnSOD and whole blood Mn levels appeared to correlate inversely. It is possible that with especially low levels of available Mn, the MnSOD protein is dysfunctional (i.e. Mn deficient) leading to overcompensation of MnSOD expression as has been found previously for Zn and the p53 tumour suppressor gene (Fanzo et al., 2001). This would, however, only be anticipated in cases of severe depletion and would not explain the cases of low whole blood Mn levels and low MnSOD expression in the present findings (Figure 6). A second, and more likely, possibility is that whole blood Mn acts as a labile store of Mn. MnSOD is one of the three primary Mn-metalloenzymes of the body (Wedler, 1994) and, in this study, circulating leucocyte expression varied greatly between individuals. This presumably is a reflection of the requirement for mitochondrial antioxidant expression of circulating leucocytes at the time of study. Blood may thus serve as a reservoir for the rapid provision of Mn to the enzyme, hence explaining a general inverse relationship except when levels of both are low (i.e. possible Mn depletion). In terms of estimated habitual dietary Mn intake, there was no obvious difference between subjects with low MnSOD expression plus low Mn levels and the other subjects. As noted above, MnSOD expression is potentially affected by a variety of stimuli; hence, clearly further work is required to see if Mn supplementation would alter either of these parameters in subjects with low expression of MnSOD and low levels of blood Mn. Thus, while the relationship between leucocyte MnSOD expression and whole blood Mn, and even a possible relationship between cAP-P expression and dietary Mn, may provide some clues in Mn homeostasis or Mn status, these findings require confirmation in a larger number of subjects and/or longitudinal intervention studies.

In conclusion, tea drinking appears to provide substantial levels of dietary Mn in habitual moderate to heavy black-tea drinkers. This does not appear to markedly alter circulating Mn levels or expression of leucocyte MnSOD or cAP-P. However, an inverse relationship was noted between whole blood Mn and leucocyte MnSOD expression as well as a possible direct association between Mn intake and cAP-P expression; hence, further work should consider this when seeking markers of Mn status or investigating Mn metabolism in humans. The use of enzyme activity and longitudinal intervention studies may prove useful additional indicators. Black tea does appear to be a major source of dietary Mn and the USA or EU upper limits of 10–11 mg/day for Mn could be reconsidered, especially for tea drinkers.



  1. Clegg MS, Lonnerdal B, Hurley LS, Keen CL (1986). Analysis of whole blood manganese by flameless atomic absorption spectrophotometry and its use as an indicator of manganese status in animals. Anal Biochem 15, 12–18. | Article |
  2. Cottrell GS, Hooper NM, Turner AJ (2000). Cloning, expression, and characterization of human cytosolic aminopeptidase P: a single manganese(II)-dependent enzyme. Biochemistry 39, 15121–15128. | Article | PubMed | ISI | ChemPort |
  3. Davis CD, Greger JL (1992). Longitudinal changes of manganese-dependent superoxide dismutase and other indexes of manganese and iron status in women. Am J Clin Nutr 55, 747–752. | PubMed | ISI | ChemPort |
  4. Fanzo JC, Reaves SK, Cui L, Zhu L, Wu JY, Wang YR et al. (2001). Zinc status affects p53, gadd45, and c-fos expression and caspase-3 activity in human bronchial epithelial cells. Am J Physiol Cell Physiol 28, 751–757.
  5. Finch S, Doyle W, Lowe C, Bates CJ, Prentice A, Smithers G et al. (1998). Report of the Diet and Nutrition Survey. National Diet and Nutrition Survey People Aged 65 and Over. London: The Stationery Office. pp. 318–320.
  6. Finley JW, Penland JG, Petit RE, Davis CD (2003). Dietary manganese intake and type of lipid do not affect clinical or neuropsychological measures in healthy young women. J Nutr 133, 2849–2856. | PubMed | ISI | ChemPort |
  7. Fraile AL, Flynn A (1991). Absorption of manganese from tea in suckling rats. Proc Nutr Soc 50, 114.
  8. Freeland-Graves JH, Behmardi F, Bales CW, Dougherty V, Lin PH, Crosby JB et al. (1988). Metabolic balance of manganese in young men consuming diets containing five levels of dietary manganese. J Nutr 118, 764–773. | PubMed | ChemPort |
  9. Gillies ME, Birkbeck JA (1983). Tea and coffee as sources of some minerals in the New Zealand diet. Am J Clin Nutr 38, 936–942. | PubMed | ISI | ChemPort |
  10. Greger JL (1998). Dietary standards for manganese: overlap between nutritional and toxicological studies. J Nutr 128, 368S–371S. | PubMed | ISI | ChemPort |
  11. Greger JL (1999). Nutrition versus toxicology of manganese in humans: evaluation of potential biomarkers. Neurotoxicology 20, 205–212. | PubMed | ISI | ChemPort |
  12. Greger JL, Davis CD, Suttie JW, Lyle BJ (1990). Intake, serum concentrations, and urinary excretion of manganese by adult males. Am J Clin Nutr 51, 457–461. | PubMed | ISI | ChemPort |
  13. Gregory J, Lowe S, Bates CJ, Prentice A, Jackson LV, Smithers G et al. (2000). Report of the Diet and Nutrition Survey. National Diet and Nutrition Survey: Young People Aged 4–18 Years. London: The Stationary Office. pp. 313–313.
  14. Henderson L, Irving K, Gregory J, Bates CJ, Prentice A, Perks J et al. (2003a). Vitamin and Mineral Intake and Urinary Analytes. The National Diet and Nutrition Survey: Adults Aged 19–64 Years. London: The Stationary office. pp. 114–115.
  15. Henderson L, Gregory J, Irving K, Swan G, Farron M (2003b). Energy, Protein, Carbohydrate, Fat and Alcohol Intake. The National Diet and Nutrition Survey: Adults Aged 19–64 years. London: The Stationary office.
  16. Holland B, Welch AA, Unwin ID, Buss DH, Paul AA, Southgate DAT (1991). McCance and Widdowson's the Composition of Foods, 5th edn. London: Royal Society of Chemistry/MAFF.
  17. Hope S-J (2005). Human Exposure to Manganese: a Dietary Study. PhD Thesis. London: Imperial College.
  18. IDA (1997). Integrated Dietary Analysis. London: IDA Publications Ltd.
  19. International Tea Committee (1995). Ann Bull Stat, pp. 123.
  20. IOM (2001). Manganese. In: Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium and Zinc. National Academy of Sciences, Washington DC: National Academic Press. pp. 394–419.
  21. Kulkarni GV, Deobagkar DD (2002). A cytosolic form of aminopeptidase P from Drosophila melanogaster: molecular cloning and characterization. J Biochem (Tokyo) 131, 445–452. | PubMed | ChemPort |
  22. Luna AS, de Campos RC (1999). Determination of Mn in whole blood and urine by graphite furnace AAS using different modifers. Atom Spectrosc 20, 108–112. | ISI | ChemPort |
  23. Macmillan-Crow LA, Cruthirds DL (2001). Invited review: manganese superoxide dismutase in disease. Free Radic Res 34, 325–336. | PubMed | ISI | ChemPort |
  24. MAFF (2001). Food Portion Sizes. Norwich: HMSO.
  25. Malecki EA, Huttner DL, Greger JL (1994). Manganese status, gut endogenous losses of manganese, and antioxidant enzyme activity in rats fed varying levels of manganese and fat. Biol Trace Elem Res 42, 17–29. | PubMed | ISI | ChemPort |
  26. Matsushima F, Meshitsuka S, Nose T (1993). Contents of aluminum and manganese in tea leaves and tea infusions. Nippon Eiseigaku Zasshi 48, 864–872. | PubMed | ChemPort |
  27. Matsuura H, Hokura A, Katsuki F, Itoh A, Haraguchi H (2001). Multielement determination and speciation of major-to-trace elements in black tea leaves by ICP-AES and ICP-MS with the aid of size exclusion chromatography. Anal Sci 17, 391–398. | Article | PubMed | ISI | ChemPort |
  28. Nelson M, Atkinson M, Meyer J (1997). Food portion sizes. A User's Guide to the Photographic Atlas. London: MAFF Publications. pp. 72.
  29. Nelson M, Beresford SAA, Kearney JM (2004). Nutritional epidemiology. In: Gibney MJ, Margetts BM, Kearney JM, Arab L (eds.), Public Health Nutrition. London: The Nutrition Society.
  30. Powell JJ, Burden TJ, Thompson RP (1998). In vitro mineral availability from digested tea: a rich dietary source of managanese. Analyst 123, 1721–1724. | Article | PubMed | ISI | ChemPort |
  31. SCF (2000). Opinion of the Scientific Committee on Food on the Tolerable Upper Intake Level of Manganese pp. 1–11. Reports of the Scientific Committee for food.http://europe.eu.int/comm/food/fs/sc/scf/index_en.html.
  32. Schlemmer G, Welz B (1986). Palladium and magnesium nitrates, a more universal modifier for graphite furnace atomic absorption spectrometry. Spectrochim Acta 41, 1157–1165. | Article | ISI |
  33. The American Tea Society About us: trends (2002) Internet (accessed August 2004):http://www.americateasociety.com/aboutUs.shtml.
  34. The Tea Council (2003a) Plantation to cup. Internet (accessed August 2004):http://www.tea.co.uk/tp/content/seed_cup.htm.
  35. The Tea Council (2003b) Tea break. Internet (accessed August 2004):http://www.tea.co.uk/tBreak/index.htm.
  36. Thompson KH, Godin DV, Lee M (1992). Tissue antioxidant status in streptozotocin-induced diabetes in rats. Effects of dietary manganese deficiency. Biol Trace Elem Res 35, 213–224. | PubMed | ISI | ChemPort |
  37. Tsubono Y, Sasaki S, Kobayashi M, Akabane M, Tsugane S (2001). Food composition and empirical weight methods in predicting nutrient intakes from food frequency questionnaire. Ann Epidemiol 11, 213–218. | Article | PubMed | ISI | ChemPort |
  38. Ulteamate Teas. Tea Info – Tea consumption. Internet (accessed August 2004):http://www.ulteamateteas.com/tea_consumption.html.
  39. Wedler F (1994). Biochemical and nutritional role of manganese: an overview. In: Klimis-Tavantzis D (ed.), Manganese in Health and Disease. Boca Raton, FL: CRC Press. pp. 1–38.
  40. Wheeler MD, Nakagami M, Bradford BU (2001). Overexpression of manganese superoxide dismutase prevents alcohol-induced liver injury in the rat. J Biol Chem 28, 36664–36672. | Article |
  41. Wrobel K, Wrobel K, Urbina EM (2000). Determination of total aluminum, chromium, copper, iron, manganese, and nickel and their fractions leached to the infusions of black tea, green tea, Hibiscus sabdariffa, and Ilex paraguariensis (mate) by ETA-AAS. Biol Trace Elem Res 78, 271–280. | Article | PubMed | ISI | ChemPort |


We gratefully acknowledge the support of the EPSRC through the EPSRC PTP initiative with WRc-NSF and London University (S-JH) as well as the UK Technical Tea Trade Association for financial support of the running costs. We are also grateful to Professor Sir Richard Thompson for his continuing advice and support and to Dr Peter Milligan for statistical advice and Drs Sylvaine Bruggraber, Dora Pereira and Chris Thane for helpful comments on the manuscript. We also thank the MRC London Iron Metabolism Group, especially Dr Kaila Srai and his research team, for advice on and use of the Real Time PCR. None of the authors had any affiliation with, or financial or personal interest in, the organizations sponsoring the research.



These links to content published by NPG are automatically generated


Possible protective effect of green tea intake on risk of adult leukaemia

British Journal of Cancer Scientific Correspondence

Possible protective effect of green tea intake on risk of adult leukaemia

British Journal of Cancer Scientific Correspondence

Extra navigation