Introduction
Introduction
Simple phenols, polyphenols and tannins (PPT) have been of great interest for many
years, in part because of their impact on the colour, odour and flavour of foods and
beverages [1], but more recently because of the possibility that these substances may have health-protecting
properties [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13]. PPT may be classified in several ways, for example, by biosynthetic origin, occurrence,
function or effect, or structure [1], [12], [14]. A classification based on structure and function will be used in this paper [15]. Simple phenols are substances containing only one aromatic ring and bearing at
least one phenolic hydroxy group and possibly other substituents, whereas polyphenols
contain more than one such aromatic ring. Phenols and polyphenols may occur as unconjugated
aglycones or as conjugates, frequently with sugars, organic acids, amino acids, lipids,
etc. [16].
The Diversity of Dietary Phenols, Polyphenols and Tannins
The Diversity of Dietary Phenols, Polyphenols and Tannins
The commonest simple phenols are cinnamates that have a C6-C3 structure [17], [18] accompanied by C6-C2 and C6-C1 compounds, and a few unsubstituted phenols [19], [20], [21]. In general these occur as conjugates. Flavonoids are the most extensively studied
polyphenols, all characterised by a C6-C3–-C6 structure, subdivided by the nature of the C3 element into anthocyanins, chalcones, dihydrochalcones, flavanols, flavanones, flavones,
flavonols, isoflavones and proanthocyanidins. The flavanols and proanthocyanidins
generally occur unconjugated but the others normally occur as glycosides. Since the
seminal paper of Hertog et al. [22] there has been a tendency to think of dietary PPT as encompassing only the flavonoids,
and the flavonoids per se to consist only of the three flavonols and two flavones that featured in that study,
but this is misleading and was never intended. It is not possible to say with precision
just how many individual PPT occur regularly in human diets, but on present evidence
a figure in excess of 200 seems reasonable [16].
The term ”tannin” refers historically to crude plant preparations that are capable
of converting hides to leather [23] and such preparations are not consumed as human food. However, the functional polyphenols
contained therein at high concentrations may also occur in certain foods and beverages
but at comparatively low concentrations that would render them totally ineffective
for producing leather. These polyphenols may be subdivided into the flavonoid-derived
proanthocyanidins (condensed tannins) [24] and the gallic acid-derived and ellagic acid-derived hydrolysable tannins, this
latter subgroup being of more restricted occurrence in human food (but commoner in
some animal feeds) [25]. The phloroglucinol-derived phlorotannins, while never used for preparing leather,
also have a limited occurrence in human food [19]. The more recent term ”phytoestrogen” refers to substances with oestrogenic and/or
anti-androgenic activity at least in vitro, and encompasses some isoflavones, some stilbenes, some lignans and some coumarins
[26]. The lignans are not oestrogen-active until transformed by the gut microflora [26], [27]. ”Antioxidants” is a third function-based term much used to describe PPT, but individual
compounds differ markedly in their ability to scavenge reactive oxygen species and
reactive nitrogen species, and inhibit oxidative enzymes. Mammalian metabolites of
PPT do not necessarily retain fully the antioxidant ability of the PPT found in plants
and especially not that of their aglycones as commonly tested in vitro [28], [29].
The PPT discussed above are substances found in healthy and intact plant tissues,
and in the main are of known structure. However, many traditional foods and beverages
as consumed have been produced by more or less extensive processing of such plant
tissues, resulting in biochemical or chemical transformations of the naturally-occurring
PPT. In some cases, black tea, matured red wines, and coffee beverage, for example,
these transformations may be substantial, generating large quantities of substances
not found in the original plant material. Despite significant advances in the last
decade [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], the structures of the majority of these novel compounds have yet to be elucidated.
Although often described as tannins, these substances are not functional tanning agents,
and should be referred to collectively as derived polyphenols until such time as their
full structural characterisation permits a more precise nomenclature.
The Consumption of PPT
The Consumption of PPT
There have been several attempts to estimate the quantities of PPT consumed, either
by using diet diaries or food frequency questionnaires and data on the typical composition
of individual commodities [3], [8], [42], [43], [44], [45], [46], [47], [48], [49], [50], or by diet analysis [6], 22], [51]. In comparison with the comprehensive databases providing the content in the diet
of the established micro- and macro-nutrients, data for the contents of PPT are much
more limited. Those data available for PPT content have been obtained by many different
methods of analysis, rarely take account of the effects of agricultural practice,
season, cooking or commercial processing, are not necessarily just for the edible
portion, and may be for varieties of fruit and vegetable different from those consumed
in a particular diet under investigation [45], [46], [52].
These are potentially serious limitations since quantitatively cultivars may differ
substantially in composition, and the non-edible parts of fruits and vegetables may
differ greatly both quantitatively and qualitatively, compared with the flesh or juice
[53]. In addition, domestic cooking and commercial processing may in some cases cause
extensive leaching and destruction [54], [55], [56], [57], [58], [59], [60].
Data based upon analysis of particular diets avoid these limitations but are usually
restricted to a few PPT because of the difficulties and cost associated with quantifying
so many individual compounds of known structure, to say nothing of the serious difficulties
associated with quantifying the uncharacterised derived polyphenols [61]. When such data are available, they are usually for PPT as aglycones released by
hydrolysis (to simplify the analysis still further) and generally for the flavonols
and flavones first studied by Hertog et al. [22] since these are amongst the easiest to determine [44], [47], [48], [51], [52], [62], [63], [64], [65], [66], [67], [68]. There are more limited data for flavanones [44], [64] and isoflavones after hydrolysis [44], [52], and flavanols, and proanthocyanidins [6], [69] (which occur as aglycones).
In an attempt at modest cost to shed more light on the nature and quantity of PPT
consumed by various populations we developed a computerised spreadsheet database beginning
with the several hundred papers identified in the NEODIET reviews [16] with continued updating as more information has been published or found [17], [52], [70], [71], [72], [73], [74], [75]. The database covers 80 commodities, including five alcoholic beverages, six fruit
juices and three other non-alcoholic beverages, and 14 PPT subgroups including derived
polyphenols, with every entry labelled to show the paper(s) from which the information
was taken. Data sources have been restricted to papers using specific methods of analysis:
data for ”total phenols by Folin-Ciocalteu” or ”total antioxidant power as gallic
acid equivalents” and similar, have been excluded. In order to reflect the variability
in composition, data have always been entered as ”high” and ”low” values (mean ± 2σ
wherever possible) and an overall mean for each commodity used in determining the
amounts of PPT consumed.
While recognising the limitations (discussed above) of such an approach to estimating
diet composition and the intake of PPT, using this database in conjunction with diet
diaries available from our other studies [76], [77], [78] has produced interesting data (Table [1]) and insights.
From Table [1] it is clear that PPT intakes may vary substantially, and that the flavones and flavonols,
upon which most emphasis has so far been placed [22], [44], [47], [48], [51], [52], [62], [63], [64], [65], [66], [67], [68], form a comparatively small part of the total intake for the populations studied.
The relatively low consumption of chalcones and dihydrochalcones, isoflavones, anthocyanins,
and stilbenes reflects the comparatively low consumption of apples and ciders, soya
products, dark berries and red wines by these populations. The significant contributions
made by the hydroxycinnamates (in these populations primarily reflecting coffee consumption
[17], [18]) and derived polyphenols (in these populations primarily reflecting black tea consumption
[79], [80], [81]) are striking. In this context ”black tea” refers to the beverage prepared from
the fermented leaf (as distinct from green tea) and not to the addition or otherwise
of milk to the beverage prior to consumption. This domination by PPT from black tea
and coffee indicates the importance also of considering the hydroxycinnamates and
derived polyphenols whenever assessing the dietary significance of PPT, and clearly
shows the limitations of looking only at flavonols and flavones after hydrolysis no
matter how precise per se the data for these aglycones might be.
It is important to stress that data for the composition of black tea and coffee beverage
reflect exactly what is consumed (with the exception of the dregs left in the cup)
since all transformations associated with processing and domestic preparation have
already taken place. Moreover, the NEODIET database is replete with analytical data
from numerous sources for the composition of these beverages (thus better avoiding
extreme values associated with any peculiarity of the material analysed or method
of analysis) compared with data for many fruits and vegetables. Accordingly, the estimated
consumption figures obtained using this database are likely to be more accurate than
would have been the case if solid foods were the major sources of PPT, and data were
for raw rather than after cooking/processing. This argument applies also to the data
for PPT delivered by wines and juices. Using this approach has led us to estimate
typical mean intakes of PPT for the two populations so far studied to be in the range
450 to 600 mg as aglycones.
Table 1 Mean dietary intakes of 14 classes of PPT as determined from diet-diaries and a food
composition data base
| PPT |
103 UK females aged 20 - 30 yearsa
|
50 UK males aged 27 - 57 yearsb
|
|
Estimated as conjugates |
Estimated as aglyconesc
|
Estimated as conjugates |
Estimated as aglyconesc
|
| Total, range |
100 - 2 300 |
|
30 - 2 200 |
|
| Total, mean |
780 |
451 |
1058 |
598 |
| Hydroxybenzoates |
15 |
|
23 |
|
| Hydroxycinnamates |
353 |
176 |
670 |
335 |
| Total flavonoids |
210 |
105 |
205 |
103 |
| Anthocyanins |
5 |
|
9 |
|
| Chalcones and dihydrochalcones |
0.7 |
|
|
|
| Flavanols |
64 |
|
58 |
|
| Flavanones |
22 |
|
89 |
|
| Flavones |
72 |
|
17 |
|
| Flavonols |
35 |
|
26 |
|
| Isoflavones |
9 |
|
0.13 |
|
| Proanthocyanidins |
7 |
|
6 |
|
| Ellagitannins |
23 |
|
|
|
| Derived polyphenols |
170 |
170 |
160 |
160 |
| Stilbenes |
9 |
|
|
|
| Lignans |
0.04 |
|
|
|
|
a Ref. [49]. |
|
b Ref. [50]. |
|
c Aglycones are estimated approximately by taking rutin as a representative flavonoid
and 5-caffeoylquinic acid as a representative hydroxycinnamate and adjusting for the
relevant molecular masses. |
Absorption and Metabolism of PPT
Absorption and Metabolism of PPT
Extensive studies in humans and animals have indicated that some PPT can be absorbed
in the small intestine, for example, certain cinnamate conjugates [82], [83], flavanols [84] (that occur naturally as aglycones), and certain flavonoid glucosides [85], [86] (but not the corresponding flavonoid rutinosides [87]). The mechanisms of absorption have not been completely elucidated but involve inter alia interaction of certain glucosides with the active sugar transporter (SGlT1) and lumenal
lactase-phloridzin hydrolase, passive diffusion of the more hydrophobic aglycones,
and interaction with cytosolic β-glucosidase. Although varying with PPT subclass and
matrix, when expressed relative to the total intake of PPT, only some 5 to 10 % of
the amount consumed is absorbed at this site. The major part of that absorbed (90
to 95 % for every substance so far studied) enters the circulation as mammalian conjugates
produced by a combination of methylation, sulphate conjugation, glucuronide conjugation
and, in the case of some phenolic acids, also by glycine conjugation [29]. Only a very small amount of the total PPT consumed, maximally 5 to 10 %, enters
the plasma unchanged.
The 90 to 95 % of the total PPT ingested, plus any mammalian glucuronides excreted
through the bile, pass to the colon where they are metabolised by the gut microflora.
Transformations may be extensive, and include the removal of sugars, removal of phenolic
hydroxyl groups, fission of aromatic rings, and metabolism to carbon dioxide, possibly
via oxaloacetate [88]. A substantial range of microbial metabolites has been identified, including phenols
and aromatic/phenolic acids/lactones possessing 0, 1 or 2 phenolic hydroxyl groups
and up to five carbons in the side chain [89], [90], [91], [92], [93], [94], [95], [96], [97], [98], [99], [100], [101], [102], [103], [104], [105], [106], [107], [108], [109], [110], [111]. Eubacterium is of particular interest since this species not only metabolises dietary (poly)phenols
[99], [100], [105], [106], [112], [113], [114], [115], [116], but also produces butyrate [117], a preferred energy source for colonic epithelial cells thought to play an important
role in maintaining colon health in humans. The yield of phenolic/aromatic acids is
variable (up to × 10) between individuals, but can be substantial (up to 50 %) relative
to the intake of PPT substrates [95], [103], [104], [108], [109], [110].
There is evidence from cell culture studies that some of the aromatic/phenolic acids,
e. g., benzoic, salicylic, p-coumaric and ferulic acids, are transported actively by the monocarboxylate transporter
MCT1 [118], [119], [120], [121], [122]. A comparatively small percentage of these microbial metabolites may eventually
appear unchanged in plasma or urine but the majority is subject to mammalian conjugation
as described for intact PPT.
Table [2] summarises in a semi-quantitative manner so far as current knowledge allows the
fate of a ”typical” daily consumption of some 450 to 600 mg of PPT (as aglycones)
previously defined in Table [1].
Table 2 Fate of ingested PPT
|
Aglycones (mg) |
| Estimated mean daily consumption (from Table [1]) |
450 - 600 |
∼ 5 - 10 % of intake absorbed in duodenum and excreted in urine. Of this
5 - 10 % unchanged plant (poly)phenols, and
90 - 95 % mammalian conjugates |
22 - 60
< 6
20 - 55 |
∼ 90-95 % fermented in colon (unabsorbed PPT+ enteric and entero-hepatic cycling of
glucuronides, etc.)
Poorly-defined and very variable portion (5 to 50 %?) absorbed depending on individual’s
flora and substrates.
Mainly mammalian conjugates of microbial metabolites |
400 - 570
20 - 285 |
Plasma Pseudo-Pharmacokinetics
Plasma Pseudo-Pharmacokinetics
Since for the majority of dietary PPT, neither the conjugates consumed, nor their
free aglycones, are detectable in plasma, it is rarely possible to perform true pharmacokinetic
analyses. Most so-called pharmacokinetic data that have been published relate to the
concentrations of aglycones released after hydrolysis of mammalian conjugates in plasma
or urine with commercial β-glucuronidase and/or sulphatase, and the data so obtained
are better referred to as pseudo-pharmacokinetics. Published data are summarised in
Table [3]. Although the maximum concentration achieved transiently varies to some extent with
PPT subclass and matrix in which consumed, it is unlikely that plasma metabolite concentrations
will routinely exceed 10 μM in total, and approximately 1 μM for total aglycones.
The reported T
max values range from 1 to 2.5 hours for substances absorbed in the duodenum [85], [86], [87], [123], [124], [125], [126], [127], [128], [129], [130], up to 5 to 12 hours when microbial metabolism is a prerequisite [87], [104], [131]. Elimination half-lives are very variable, ranging from as low as 1 hour [132], [133] to values in excess of 20 hours [85], [6], [87]. The very low values may be artefacts of observation periods being less than the
true half-life, whereas the very high values may be exaggerated because of a biphasic
elimination reflecting significant entero-hepatic circulation of glucuronides. When
studied explicitly, repeat dosing has failed to provide evidence of accumulation in
plasma suggesting that, in general, significant elimination occurs in a time period
shorter than the interval between repeat doses [123].
Table [4] summarises the concentrations of a range of endogenous (i. e., non-dietary) simple
phenols, including α-tocopherol, and ascorbate in plasma from healthy individuals.
The total simple phenol and ascorbate concentration is between 159 and 380 μM. The
maximum additional concentration that is likely to be achieved from dietary sources,
3 to 22 μM, is marginal by comparison adding only between 0.3 and 5 % if it is assumed,
quite reasonably, that the ”typical” mean intake is taken over three equal meals.
Many people consume a much smaller quantity of dietary PPT and even those consuming
double the average amount (450 to 600 mg aglycones) adopted in this paper will only
achieve a transient 5 to10 % increase in total plasma antioxidant content.
Many investigators have attempted to demonstrate increases in plasma antioxidant capability
following the consumption of foods, beverages or supplements rich in PPT. Table [5] summarises the outcomes of 34 such studies [127], [130], [131], [132], [133], [134], [135], [136], [137], [138], [139], [140], [141], [142], [143], [144], [145], [146], [147], [148], [149], [150], [151], [152], [153], [154], [155], [156], [157], [158], [159]. The test substances included a range of fruit and vegetable products, including
juices, alcoholic beverages, tea, and chocolate. In view of the calculations presented
in Tables [3] and 4, it is perhaps not surprising that increases in plasma antioxidant capacity were
often undetectable, and at best, small and transient. Moreover, in four studies that
produced increases in plasma antioxidant capability it could be attributed, at least
in part, to increased plasma ascorbate [149], [156], [158].
In view of these observations, it is instructive also to consider the redox potentials
of PPT-derived mammalian metabolites that are known to reach plasma, and to compare
these with the corresponding values for the endogenous plasma antioxidants. The polyphenols
with the lowest redox potentials are flavonoids with vicinal hydroxyl groups in the
B-ring, and conjugation extending to the A-ring, e. g., quercetin aglycone (330 mV
at pH 7) [160]. If the conjugation does not extend beyond the B-ring, then the redox potential
is significantly higher even for (-)-epigallocatechin gallate (480 mV at pH 7) [161] with three vicinal hydroxyl groups. The value rises again when there are only two
vicinal hydroxyl groups {e. g., (+)-catechin 570 mV [162] or caffeic acid 540 mV [162]}, a single para hydroxyl group (e. g., hesperidin 720 mV [162]) or isolated (meta) hydroxyl groups (e. g., resorcinol 810 mV [163]). These comparisons are extended to the endogenous (non-dietary) plasma antioxidants
in Table [6]. Fig. [1] illustrates the marked effects of mammalian and microbial metabolism on the redox
potential of PPT aglycones that are frequently examined in in vitro systems designed to demonstrate their potent antioxidant properties. When the aglycones
of such powerful antioxidants are given intravenously to humans [164] or intraperitoneally to animals [165], [166], thus circumventing the protection offered by the gastric mucosa and Phase II conjugations,
redox cycling causes serious and possibly fatal liver and kidney damage. One may conclude
that it is better to avoid high plasma concentrations of the more potent PPT antioxidants
(such as unconjugated quercetin), and that it might be ill-advised grossly to supplement
normal diets with capsules and concentrates of such potent antioxidants.
Table [6] indicates that the diet-derived PPT metabolites are able thermodynamically to scavenge
some or all of the damaging radicals should they come into contact. However, these
metabolites are so hydrophilic {e. g., quercetin 3-glucuronide (K = 0.008) [167], [168] compared with quercetin (K = 66) [167], [168] and α-tocopherol (K = 550) [169]} that it is unlikely they will encounter lipid-derived radicals. However, any phenoxyl
radicals generated will have to be removed either by transfer of the unpaired electron
to an endogenous scavenger such as α-tocopherol, ascorbate, glutathione or serum albumin,
or by dismutation or disproportionation although these latter mechanisms seem somewhat
unlikely in vivo because of the relatively low phenoxyl radical concentrations. The implied demand
for α-tocopherol and ascorbate is particularly interesting, since two of the supplementation
studies (Table [5]) produced reductions in plasma ascorbate and α-tocopherol [150], and the major sources of dietary PPT identified from the NEODIET database (coffee
and black tea) supply neither. Moreover, it is known that for approximately 14 % of
the over-65 population subgroup in the UK the mean plasma ascorbate value is below
11 μM [170], indicating biochemical depletion [171], suggesting that for heavy consumers of black tea or coffee within this population
subgroup the transient concentration of PPT metabolites may approach or even exceed
plasma ascorbate.
From the data assembled, it is difficult to envisage how diet-derived PPT metabolites
can make a major contribution to radical scavenging in plasma compared with the contribution
to be expected from the endogenous antioxidants in healthy individuals replete in
ascorbate. It follows that if diets rich in fruits and vegetables are advantageous,
at least in part, by virtue of their content of PPT then mechanisms other than radical
scavenging are implicated.
Table 3 Plasma pseudo-pharmacokinetics after consumption of normal portions of rich sources
| PPT Subclass |
C
max (nM) unchangeda
|
C
max (nM) mammalian conjugates |
% urine excretion |
| Anthocyanins |
10 - 150 |
traces |
N.D. - 0.1b
|
Flavanols, low fat
Flavanols, high fat |
40 - 140
150 - 220 |
1000 - 2000
up to 6200 |
0.5 - 4.0
25 - 30 |
Flavonol glycosides
Flavonol aglycones |
Minute traces
Minute traces |
N.D.b
350 - 1100 |
0.5 - 2.5 |
| Flavanone glycosides |
Minute traces |
120 - 1500 |
4 - 10 |
Isoflavone glycosides
Isoflavone aglycones |
Minute traces
10 - 150 |
900 - 4000
500 - 6500 |
20 - 55 |
| Cinnamates & chlorogenic acids |
up to 120 |
up to 500 |
1 - 2 |
| Phenolic gut flora metabolites |
|
20 - 60 |
Up to 50 |
|
Hypothetical total if all consumed in one meal
|
250 - 780
|
2890 - 21660
|
|
|
a C
max = maximum concentration achieved transiently in plasma. |
|
b N.D. = not detected. |
Table 4 Plasma concentrations (μM) of endogenous (non-dietary) phenols and other plasma antioxidants
|
Plasma concentration,
healthy individuals |
Homogentisic acid
p-Hydroxyphenyl lactate
p-Hydroxyphenyl pyruvate
Tyrosine |
0.014 - 0.070a
40 - 90b
14 - 60b
60 - 130b,c
|
Ascorbate
α-Tocopherol |
40 - 70d,e
5 - 30f
|
|
Total endogenous phenols & antioxidants
|
159 - 380
|
Hypothetical total diet-derived phenols
Averaged over three meals gives a transient
increase of between 0.3 and 5 %.
Many people consume much less |
3.1 - 22.4g
≈ 1 - 7.5g
|
|
a Ref. [201], b Ref. [202], c Ref. [203], d Ref. [204], e Ref. [205], f Ref. [206], g From Table [3]
|
Table 5 The outcome of 34 studiesa in which volunteers were given foods, beverages or supplements rich in PPT and plasma
was analysed for total antioxidant activity
| 34 studies, (poly)phenol-rich diet compared with control |
- 13 studies (3 high & 3 very high doses) showed no change in plasma
antioxidant status ex vivo
|
- 21 studies (8 low & 7 moderate doses) showed small and transient increases
in plasma antioxidant status ex vivo
|
| - 1 showed reduction in plasma Vitamin E |
| - 1 showed reduction in plasma ascorbate and glutathione |
|
a Refs: [127], [130], [131], [132], [133], [134], [135], [136], [137], [138], [139], [140], [141], [142], [143], [144], [145], [146], [147], [148], [149], [150], [151], [152], [153], [154], [155], [156], [157], [158], [159]. |
Table 6 A summary of published data for transient maximal plasma concentrations of diet-derived
(poly)phenols, typical plasma concentrations of endogenous phenols and antioxidants,
and associated redox potentials (pH 7)
|
Mammalian metabolite hydroxylation pattern
|
Maximal transient concentration (μM)
|
Redox potential (mV) at pH 7
|
1,2,3-vic
1,2-vic
Single para or isolated meta hydroxy groups
Blocked/inactive |
0.14a
0.8a
10a
? |
400 - 600d,e,f
500 - 650d,g,h,
700 - 1050d,e,g,i,j,k
Inactive |
|
Damaging radicals
|
|
Redox potential (mV) at pH 7
|
Hydroxyl radical
Superoxide radical anion
Alkoxyl radical
Alkyl-peroxyl radical
PUFA (bis-allylic) radical |
|
2310 h
1800 h
1600 l
1000 ± 60h,l,m,
600 h
|
|
Endogenous phenols and antioxidants in plasma
|
Typical plasma concentration (μm)
|
Redox potential (mV) at pH 7
|
Endogenous p-phenols
α-Tocopherol
Ascorbate
(depleted)
Glutathione |
114, 280a
5 - 30
50 - 70b
(≤ 11)c
|
≈ 700d,e,g,i,j,k
≈ 500d,h
≈ 280e,h
- 276n
|
|
a From Table [3], b From Table [4], c Ref. [170], d Ref. [161], e Ref. [163], f Ref. [207], g Ref. [162], h Ref. [208], i Ref. [209], j Ref. [210], k Ref. [211], l Ref. [212], m Ref. [213], n Ref. [214]
|
Fig. 1 Illustration of the effects of mammalian metabolism and microbial metabolism on the
redox potential of (poly)phenols found in plasma compared with their precursors in
the diet and the aglycones commonly used in in vitro studies.
Protective Mechanisms other than Radical Scavenging
Protective Mechanisms other than Radical Scavenging
Although classically, mammalian conjugates of drugs are viewed as biologically significantly
less active than the parent drug, this is not inevitably the case when PPT are considered.
Some quercetin conjugates are able to inhibit lipoxygenase and xanthine oxidase. Quercetin
3-glucuronide, one of the three major human conjugates of dietary quercetin glycosides,
has been shown in vitro to protect the vascular endothelium [172], [173] and suppresses peroxynitrite-induced consumption of lipophilic antioxidants in human
LDL [174]. Another human metabolite, quercetin 4′-glucuronide, inhibits xanthine oxidase in vitro at a concentration in plasma that on normal diets can realistically be approached
(K
i = 0.25 μM) [175]. As these observations are more widely appreciated, and mammalian metabolites become
more readily available, it is quite possible that other biologically interesting properties
will be identified for mammalian conjugates of PPT.
Effects occurring in the gastro-intestinal tract prior to absorption also deserve
greater consideration. For example, there is a growing body of evidence suggesting
that diets rich in PPT may influence the absorption and metabolism of glucose, resulting
in a lower glycaemic index [176] than would otherwise be expected. Red wine [177], coffee [178] and apple juice [179] have all been shown in controlled volunteer studies to slow glucose absorption and
reduce the post-prandial surge in plasma glucose, an event known to be an independent
risk factor for CHD [180]. Studies in which volunteers consumed normal portions of PPT-rich foods [178] have also produced reductions in the post-prandial concentrations of plasma insulin
and glucose-dependent insulinotropic polypeptide and elevation in the concentration
of glucagon-like polypeptide-1. A prospective study of 17,000 people suggested that
the mean relative risk of developing Type II diabetes was only 0.5 (0.35 - 0.72) in
those individuals habitually consuming six or more cups of coffee per day compared
with those consuming two or less (p = 0.0002) [181], and a polyphenol-enriched diet has been reported to reduce the incidence and severity
of nephropathy in Type II diabetics [182].
The reduced glycaemic index has been attributed to PPT-mediated inhibition of α-amylase
[183], [184], maltase [185] or α-glucosidase (sucrase) [184], [186], but this mechanism would not operate with preformed glucose as observed in our
studies [178], [179]. In studies using bolus doses of glucose, the observation is more conveniently explained
by an effect on the active glucose transporter (SGLT1) in the duodenum. Phloridzin,
a dihydrochalcone glucoside characteristic of apples and apple products [53], but now known to be more widely distributed [187], competes for the active site both in vitro and in vivo when given intraperitoneally [188], [189], [190], [191]. Other dietary PPT [(-)-epigallocatechin gallate, (-)-epigallocatechin and 5-caffeoylquinic
acid] have been shown in vitro to dissipate the sodium gradient essential to the operation of SGLT1 [192], [193], and several quercetin glucosides have been shown to interact with it and thus to
have the potential to interfere in glucose transport [194], [195], [196], [197], [198], [199], [200]. While these effects on glucose absorption and the associated hormones are modest,
they have been achieved in volunteers consuming sensible quantities of common dietary
components (as distinct from effects seen only in vitro with high levels of PPT aglycones never seen in the diet). Such effects repeated
daily, or even several times daily for say 30 years, might in part explain the reduced
incidence of chronic disease, especially Type II diabetes and the metabolic syndrome,
in later life associated with diets rich in fruits and vegetables.
Conclusions
Conclusions
Evidence has been presented to indicate that very little of the dietary PPT consumed
reaches human plasma, and that this transient fraction contains only weak antioxidants
able to make little contribution to the total antioxidant activity of the plasma.
This suggests that radical scavenging by PPT is unlikely to be the mechanism by which
diets rich in fruits and vegetables protect against chronic diseases. Instead, it
is proposed that modulation of glucose absorption in the duodenum prior to the absorption
of PPT, with protection by PPT metabolites through mechanisms other than radical scavenging,
might over a lifetime offer modest protection against chronic diseases, especially
Type II diabetes and the metabolic syndrome.
