Abstract
Background: The incidence of many diseases increases with age and leads
to multimorbidity, characterized by the presence of multiple diseases
in old age. This phenomenon is closely related to systemic metabolic
changes; the most suitable way to study it is through metabolomics. The
use of accumulated metabolomic data to characterize this phenomenon at
the system level may provide additional insight into the nature and
strength of aging–disease relationships. Methods: For this purpose,
metabolic changes associated with human aging and metabolic alterations
under different pathological conditions were compared. To do this, the
published results of metabolomic studies on human aging were compared
with data on metabolite alterations collected in the human metabolome
database through metabolite set enrichment analysis (MSEA) and
combinatorial analysis. Results: It was found that human aging and
pathological conditions involve the set of the same metabolic pathways
with a probability of 99.96%. These data show the high identity of the
aging process and the development of diseases at the metabolic level
and allow to identify the set of metabolic pathways reflecting
age-related changes closely associated with health. Based on these
pathways, a metapathway was compiled, changes in which are
simultaneously associated with health and age. Conclusions: The
knowledge about the strength of the convergence of aging and
pathological conditions has been supplemented by the rigor evidence at
the metabolome level, which also made it possible to outline the age
and health-relevant place in the human metabolism.
Keywords: metabolomics, aging, pathological conditions
1. Introduction
Simultaneously with a dropping birth rate, the lifespan of people
worldwide has been continuously rising in recent decades [[32]1].
Population aging is now a widespread phenomenon and one of the most
significant social changes in the twenty-first century, posing economic
and social challenges, particularly in the area of health care, given
that aging is often accompanied by cardiovascular, respiratory, and
mental diseases, arthritis, diabetes, and disability [[33]2,[34]3]. To
find ways to reduce aging symptoms, researchers continue to explore the
molecular mechanisms of aging [[35]4,[36]5].
Aging causes the body to undergo numerous changes at all organizational
levels, from genome to metabolome. However, there is still much to be
understood about the molecular basis of aging. Over the past 30 years,
gerontological research has made significant strides in understanding
how genes regulate aging [[37]6,[38]7]. Genomic studies place a lot of
emphasis on the variables that influence lifespan and successful aging,
defined as the absence of chronic diseases and the capacity to act well
at physiological and psychological levels [[39]8,[40]9,[41]10,[42]11].
However, the intricate combination of multiple factors (ranging from
hereditary to numerous environmental ones) determines a lifespan
[[43]12]. Transcriptomics, proteomics, and metabolomics are examples of
fundamental postgenomic omics sciences that can reveal further details
regarding alterations in the organism at “post-genome” molecular levels
[[44]13,[45]14,[46]15]. Metabolomics holds a unique position in
scientific research since the metabolome serves as the culmination
point for biological events that emerge from complex interactions
between genes, proteins, biochemical processes, and environmental
factors [[47]16,[48]17,[49]18]. Metabolites, low-molecular compounds
that are substrates, intermediates, and products of metabolic events
taking place in the body, form metabolomes [[50]4,[51]19]. As a result,
the metabolome analysis can offer details on the body’s current
metabolic status in relation to physiological and pathological
processes [[52]20,[53]21] as well as reveal the metabolic signatures of
organism aging [[54]22,[55]23,[56]24].
For the global metabolome overview, untargeted metabolomics approaches
measuring bid sets of metabolites are ideally suitable
[[57]22,[58]25,[59]26,[60]27,[61]28,[62]29,[63]30]. In a previous
study, a comparative analysis of untargeted metabolomics aging studies
in various animal models (from C. elegans to mammals) and humans was
conducted using metabolite set enrichment analysis (MSEA), which showed
the identity of aging among them from a metabolomics point of view
[[64]31]. This result stimulated further study of the aging process
using the same approach. The purpose of this work is to compare
metabolic changes associated with human aging with metabolic
alterations defined in pathological conditions. To do this, the
published results of untargeted metabolomic studies on human aging were
compared with data on human metabolite alterations at different
pathological conditions collected in the human metabolome database
([65]Figure 1).
Figure 1.
[66]Figure 1
[67]Open in a new tab
Workflow for comparison of metabolomic data collected on human aging
and pathological conditions. p-value indicates a probability that the
lists of human aging-related pathways and pathways associated with
pathological conditions are the same. MSEA, metabolite set enrichment
analysis.
2. Materials and Methods
2.1. Metabolite Datasets
Studies focusing on untargeted metabolomic analysis of the blood
metabolome during aging [[68]32,[69]33,[70]34,[71]35,[72]36,[73]37]
have been previously reviewed [[74]31]. The aging-associated
metabolites in humans identified in these studies were used as
metabolite dataset 1.
Metabolites associated with human pathological conditions were obtained
from the Human Metabolome Database (HMDB; [75]https://hmdb.ca; accessed
on 1 September 2022) and used as metabolite dataset 2. For this, the
database was downloaded in XML format from the HMDB website, parsed in
MATLAB program (version R2019a; MathWorks, Natick, MA, USA), and
dataset 2 with metabolites at abnormal concentrations and associated
pathological conditions was compiled.
2.2. Metabolite Set Enrichment Analysis
MSEA [[76]38] was applied using MetaboAnalyst 5.0 software
([77]www.metaboanalyst.ca; accessed on 1 September 2023) [[78]39] with
the following options: module, ‘pathway analysis’; input type,
‘compound names’ or ‘KEGG ID’; visualization method, ‘scatter plot
(testing significant features)’; enrichment method, ‘hypergeometric
test’; topology analysis, ‘relative-betweenness centrality’; reference
metabolome, ‘use all compounds in the selected pathway library’;
pathway library, ‘homo sapiens (KEGG)’ (80 pathways; KEGG pathway info
were obtained in October 2019). The enrichment analysis (type
over-representation analysis (ORA)) implemented using the
hypergeometric test evaluated whether a particular metabolite set was
represented more than expected by chance within the given compound
list. One-tailed p-values were provided by ORA after adjusting for
multiple testing.
2.3. Combinatorial Analysis
To assess whether the presence of the same metabolic pathways
associated with human aging and pathological conditions is
statistically significant, combinatorial analysis was used. Namely, the
combinatorial problem was solved: the calculation of the probability
that a defined set of metabolic pathways associated with human aging
could appear among metabolic pathways associated with pathological
conditions.
2.4. Formation and Analysis of Health and Aging-Related Metapathway
Pathways enriched both in aging and in pathological conditions were
used to compile health and aging-related metapathway. This metapathway
was used to apply enrichment analysis with the same parameters as
described in [79]Section 2.2. For enrichment analysis, the option of
MetaboAnalyst 6 for uploading custom metabolite pathways
([80]www.metaboanalyst.ca; accessed on 15 August 2024) was used to add
metapathway to default KEGG pathway-based 80 metabolite sets (KEGG
pathway info was obtained in December 2023).
Projection of metabolites, included in metapathway, onto KEGG pathways
was conducted by the network analysis module of the MetaboAnalyst 6
(option ‘KEGG global metabolic network’; accessed on 8 September 2024).
The metabolite–metabolite interaction network for metabolites included
in the metapathway was built using the same module (option
‘metabolite–metabolite interaction network’; layout
Fruchterman–Reingold). According to the module description, the
chemical–chemical associations for the metabolite network were
extracted from STITCH [[81]40], so that only highly confident
interactions are used.
3. Results
3.1. Metabolic Pathways Associated with Human Aging and Pathological
Conditions
In a previous study [[82]31], untargeted metabolomic studies of blood
have been reviewed in order to observe human metabolome changes during
aging [[83]32,[84]33,[85]34,[86]35,[87]36,[88]37]. The obtained list of
human aging-associated metabolites, including study reference, sample
type, age of involved subjects, and metabolite detection method, is
presented in [89]Supplementary Table S1. This set of aging-associated
metabolites was used as metabolite dataset 1 in this study.
In accordance with the goal of this study, which is to compare
metabolic changes linked to human aging with metabolic variations under
various pathological conditions, dataset 2 was compiled from the HMDB
using data on metabolites at abnormal concentrations under 326
pathological conditions. The resulting metabolite dataset 2 is
presented in [90]Supplementary Table S2 (list of pathological
conditions with corresponding metabolite identifiers).
Both metabolite datasets were subjected to MSEA. [91]Figure 2 and
[92]Table 1 show metabolic pathways, which are considered to be
involved in human aging and pathological conditions based on projecting
the metabolites in dataset 1 and dataset 2 onto human metabolic
pathways by MSEA.
Figure 2.
[93]Figure 2
[94]Open in a new tab
Results of metabolite set enrichment analysis (MSEA). (a) Metabolic
pathways potentially involved in human aging. The graph was generated
by projecting metabolites associated with human aging (metabolite
dataset 1) onto metabolic pathways using MSEA. (b) Metabolic pathways
potentially involved in human pathological processes. The graph was
generated by projecting metabolites for which abnormal concentrations
are described in the Human Metabolome Database (metabolite dataset 2)
onto human metabolic pathways using MSEA. The names of metabolic
pathways concurrently associated with human aging and pathological
conditions (i.e., enriched with projected metabolites with an adjusted
p-value < 0.05) are shown. The p-value, determined by the pathway
enrichment analysis, evaluates whether a measured set of metabolites is
represented in the pathway more than expected by chance within a given
list of metabolites. Pathway impact values are from the pathway
topology analysis.
Table 1.
Results of MSEA obtained by projecting metabolites associated with
human aging and with abnormal concentrations in various pathological
conditions onto human metabolic pathways.
# Pathway Name ^1 Match Status
(Hits/Total) ^2 p-Value FDR ^3
For human aging
1 Aminoacyl-tRNA biosynthesis 14/48 1.3 × 10^−7 1.1 × 10^−5
2 Arginine biosynthesis 7/14 3.7 × 10^−6 1.6 × 10^−4
3 Histidine metabolism 7/16 1.1 × 10^−5 3.2 × 10^−4
4 Alanine, aspartate, and glutamate metabolism 8/28 9.6 × 10^−5 0.0020
5 Glyoxylate and dicarboxylate metabolism 8/32 2.7 × 10^−4 0.0045
6 Valine, leucine, and isoleucine biosynthesis 4/8 5.7 × 10^−4 0.0069
7 Citrate cycle (TCA cycle) 6/20 5.8 × 10^−4 0.0069
8 Phenylalanine, tyrosine, and tryptophan biosynthesis 3/4 6.8 × 10^−4
0.0071
9 Nicotinate and nicotinamide metabolism 5/15 0.0010 0.0094
19 Phenylalanine metabolism 4/10 0.0016 0.0120
11 Caffeine metabolism 4/10 0.0016 0.0120
12 D-glutamine and D-glutamate metabolism 3/6 0.0031 0.0209
13 Biosynthesis of unsaturated fatty acids 7/36 0.0032 0.0209
14 Ascorbate and aldarate metabolism 3/8 0.0080 0.0458
15 Butanoate metabolism 4/15 0.0082 0.0458
Metapathway (6 united pathways) 23/76 3.6 × 10^−12 2.5 × 10^−10
For abnormal concentrations in various pathological conditions
1 Aminoacyl-tRNA biosynthesis ✓ 20/48 6.3 × 10^−9 5.3 × 10^−7
2 Glycine, serine and threonine metabolism 15/33 1.4 × 10^−7 6.0 ×
10^−6
3 Arginine biosynthesis ✓ 8/14 1.7 × 10^−5 4.8 × 10^−4
4 Valine, leucine, and isoleucine biosynthesis ✓ 6/8 2.4 × 10^−5 5.1 ×
10^−4
4 Steroid hormone biosynthesis 21/85 5.6 × 10^−5 9.5 × 10^−4
6 Alanine, aspartate and glutamate metabolism ✓ 10/28 2.3 × 10^−4
0.0033
7 Tyrosine metabolism 12/42 6.1 × 10^−4 0.0073
8 Butanoate metabolism ✓ 6/15 0.0023 0.0246
9 Glyoxylate and dicarboxylate metabolism ✓ 9/32 0.0034 0.0290
10 Arginine and proline metabolism 10/38 0.0035 0.0290
11 Phenylalanine, tyrosine, and tryptophan biosynthesis ✓ 3/4 0.0039
0.0294
Metapathway (6 united pathways) 29/76 1.3 × 10^−10 1.1 × 10^−8
[95]Open in a new tab
^1 Only metabolic pathways with FDR < 0.05 are presented. ^2 ‘Hits’ is
the matched number of metabolites from the datasets. ‘Total’ is the
total number of compounds in the pathway. ^3 p-value adjusted using
False Discovery Rate (FDR). A checkmark (✓) indicates pathological
condition-related pathways, which are also human aging-related.
Based on the MSEA, it may be concluded that the untargeted metabolomic
data collected to date on human aging suggests that the 15 metabolic
pathways are statistically significantly involved in the aging process.
The information gathered from metabolomic studies on various metabolite
alterations and stored in the HMDB thus far indicates the statistically
significant involvement of 11 metabolic pathways. Among them, seven
metabolic pathways were found from the MSEA of metabolites from
metabolomic studies on human aging (see checkmarked pathways in
[96]Table 1).
3.2. Combinatorial Analysis Result
To confirm that the MSEA results for human aging-related metabolites
and pathological conditions describe the same biological phenomenon,
combinatorial analysis was used. The probability (p) that seven human
aging-related pathways were non-randomly found among pathological
condition-related pathways was calculated.
An elementary event—there are 15 human aging-associated metabolic
pathways. An evaluated event: out of 11 pathological condition-related
metabolic pathways, 7 are human aging-related (presuming that the
processes of aging and pathological conditions reflected in metabolic
pathways are not independent). Considering that there are only 80 human
metabolic pathways, by choosing 11 metabolic pathways out of 80, a
sample set can be obtained by combination:
[MATH:
n=C<
mn>8011=80!
11!80−11<
mo>! :MATH]
Non-aging-related pathways could appear among pathological
condition-related pathways by selecting 4 pathways out of 65 pathways,
which gives the total number of such samples:
[MATH:
m1=C654=<
mstyle scriptlevel="0"
displaystyle="true">65!
4!65−4! :MATH]
Similarly, the number of possible groups of 7 aging-related pathways is
determined by a combination of 7 out of 15:
[MATH:
m2=C157=<
mstyle scriptlevel="0"
displaystyle="true">15!
7!15−7! :MATH]
The aging-related and non-aging-related pathways fall into the set of
pathological condition-related pathways independently of each other;
therefore, to calculate the number of elementary events favorable to an
evaluated event, the multiplication rule (“and” rule) of combinatorics
was used. So, the total number of favorable elementary events:
[MATH:
m=m1×m2 :MATH]
The probability of the evaluated event was determined by the formula:
[MATH: p=1−mn=0.9996(or 99.96%) :MATH]
The resulting probability value indicates that aging-related and
pathological condition-related pathways reflect or belong to the same
biological phenomenon with a probability of 99.96%.
3.3. Health and Aging-Related Metapathway
Metabolic pathways associated with both aging and pathological
conditions ([97]Table 1), such as ‘Arginine biosynthesis’, ‘Valine,
leucine, and isoleucine biosynthesis’, ‘Alanine, aspartate, and
glutamate metabolism’, ‘Butanoate metabolism’, ‘Glyoxylate and
dicarboxylate metabolism’, ‘Phenylalanine, tyrosine, and tryptophan
biosynthesis’, except for ‘Aminoacyl-tRNA biosynthesis’, were combined
into health and aging-related metapathway. The enrichment of the
metapathway by both aging and pathological conditions-associated
metabolites was equally high (p-value of 3.6 × 10^−12 and 1.3 × 10^−10,
respectively; [98]Table 1).
The metabolite–metabolite interaction network helps to highlight
potential functional relationships between metabolites. Using data from
STITCH (‘search tool for interactions of chemicals’) [[99]40], which
integrate information about interactions from metabolic pathways,
binding experiments, and drug–target relationships, to construct an
interaction network, it is possible to construct a full-fledged
interaction network based on metabolites detected in metabolomic
studies, supplementing them with interacting but not detected
metabolites. The network also identifies metabolites with the highest
number of functional connections (degree of nodes), which determines
their importance in the network.
Such an interaction network built for metabolites of the metapathway
([100]Figure 3) shows multiple connections between them, indicating the
functional integrity of the metapathway. Metabolites with the largest
number of connections (metabolites in the center of the network) are
presented in [101]Table 2.
Figure 3.
[102]Figure 3
[103]Open in a new tab
Metabolite–metabolite interaction network for metapathway metabolites.
Circle size and color correspond to the degree number of the node.
Degree values for central nodes (metabolites), indicated as large red
circles, are presented in [104]Table 2 (data for all metabolites are
presented in [105]Supplementary Table S3).
Table 2.
Metabolite–metabolite interaction network parameters for central nodes
(metabolites).
Metabolite Name Degree Betweenness
L-Glutamic acid 57 234.89
Oxoglutaric acid 53 88.12
Adenosine triphosphate 51 179.78
Pyruvic acid 50 82.26
NADP 46 82.82
Carbon dioxide 46 80.72
NADH 44 108.28
Oxygen 40 52.65
[106]Open in a new tab
The data in [107]Table 2 indicate that the central role in the
metapathway is assigned to metabolites related to energy metabolism.
Such metabolites as ATP, NADH, NADP, carbon dioxide, oxygen, and
pyruvic acid are directly related to cellular respiration and reflect
the cellular energy state. Oxoglutaric acid, as a TCA intermediate, and
glutamic acid, which is converted from oxoglutaric acid using NADP, are
also highly relevant to energy metabolism.
4. Discussion
A useful tool for researching aging-related biological processes is
metabolome analysis [[108]16,[109]25,[110]29]. Metabolome analysis aims
to measure many substances with low molecular weights simultaneously in
biological samples [[111]41,[112]42,[113]43]. Nuclear magnetic
resonance spectroscopy [[114]44,[115]45] and mass spectrometry
[[116]46,[117]47,[118]48] provide information about large sets of
metabolites in a biological sample in a single analysis. Pico- and
femtomole concentration analysis of hundreds or thousands of
metabolites in a sample is possible with mass spectrometers [[119]49].
As a result, metabolomics allows gathering information on a wide range
of metabolites belonging to different chemical classes and metabolic
pathways.
Two main types of metabolome analysis can be identified: targeted for
measuring predefined sets of metabolites and untargeted, which is
related to panoramic detection of sample metabolites
[[120]28,[121]34,[122]50]. According to the aim and design of this
study, the results of untargeted (panoramic) studies are more suitable
since the results of targeted studies are limited to predefined sets of
metabolites, making it difficult to draw generalized conclusions.
Moreover, MSEA, being a central method in this study, distorts the
results when a predefined set of metabolites is used. A review of
untargeted metabolomics studies related to aging was conducted in a
previous study [[123]31], and the human aging data from that review
were used in this study as dataset 1.
A wide investigation of alterations in metabolic pathways was conducted
through analysis of metabolomes over the past two decades. Several
databases collect results from these studies, where metabolites are
annotated with information about their chemical composition, method of
detection, concentration data for both normal and pathological
conditions, and metabolic pathways in which they participate. HMDB is
one of the most well-known databases with detailed descriptions of
metabolites that have been identified and present in the human body.
The data on metabolites with abnormal concentrations were extracted
from this database to compile dataset 2 for comparison with dataset 1.
When considering the results of different metabolomic studies to draw
generalized conclusions, it is necessary to take into account the
specifics of such studies. The measurement of large sets of metabolites
using different samples, sample preparation protocols, and measurement
equipment leads to different sets of metabolites being measured in
untargeted metabolomic studies. This makes it difficult to generalize
findings from the different studies that are directly relevant to our
study. A review of the most frequent metabolites in dataset 1 and
dataset 2 ([124]Supplementary Figures S1 and S2) confirms that finding
a set of metabolites common to both datasets is challenging. This
problem can be solved by the projection of identified metabolites from
different studies onto metabolic pathways. If different metabolites
found in different studies are projected onto the same metabolic
pathway, this indicates their participation in identical biochemical
processes, and such data can be interpreted in the same way. MSEA,
which enables getting a statistical estimate of such projection, is one
of the popular methods for doing this [[125]51]. MSEA provides the
p-value that the measured metabolites match a certain metabolite set,
in particular a metabolic pathway ([126]Figure 4). If different
metabolites detected in different studies are involved in the same
metabolic pathway, this suggests that they are involved in the same
biochemical processes. Therefore, in this research, the projection of
datasets 1 and 2 into metabolic pathways using MSEA is the most
appropriate way to compare them.
Figure 4.
[127]Figure 4
[128]Open in a new tab
Principle of metabolite set enrichment analysis (MSEA) applied to
metabolic pathways.
Combinatorial analysis of MSEA results suggested that metabolomic data
on aging and pathological conditions have a high degree of identity. Of
the fifteen aging-related pathways, seven appeared among the eleven
pathways associated with pathological conditions. Combinatorial
analysis indicates that we observe similar processes with a probability
of 99.96%. Moreover, the similarity is even more pronounced than the
resulting probability conveys. For example, the presence of the steroid
hormone biosynthesis pathway among pathological condition-related
pathways and the absence among aging-related pathways, although the
relationship between steroids and aging is well-known, may be
explained. Several steroids collected from the aging-related studies
were not recognized and, therefore, not processed by MetaboAnalyst due
to the diversity of their structural conformations and the way their
names are spelled. So, the difference in steroids between MSEA results
is more a limitation of method than a real difference. With this
comment in mind, the difference between pathways associated with aging
and pathological conditions becomes even less defined.
Moreover, matched aging-related pathways and abnormal
metabolite-related pathways have close ranking orders. In the first
place in both cases is ‘Aminoacyl-tRNA biosynthesis.’ ‘Arginine
biosynthesis’, ‘Alanine, aspartate, and glutamate metabolism’, ‘Valine,
leucine, and isoleucine biosynthesis’ occupy an intermediate position.
‘Phenylalanine, tyrosine, and tryptophan biosynthesis’ and ‘Butanoate
metabolism’ are towards the end of the list. This pattern is another
sign in favor of the identity of MSEA results for dataset 1 and dataset
2. If a scatter plot of these metabolic pathways is built ([129]Figure
5), then they can be linearly approximated with a coefficient of
determination (R^2) equal to 0.78, which objectively confirms the
presence of similarity in their ranking for pathological conditions and
aging.
Figure 5.
[130]Figure 5
[131]Open in a new tab
Scatter plot of human aging-associated metabolic pathways versus
pathological condition-associated pathways. The p-value evaluates
whether a measured set of metabolites is represented in the pathway
more than expected by chance within a given list of metabolites
(p-values from the pathway enrichment analysis). R^2 is a coefficient
of determination calculated for the linear approximation of scatter
plot points.
The obtained results allow us to state that the development of
pathological conditions and aging have great similarities at the
metabolic level, which, in some way, allows us to assume that they are
related to the same general biological process. It is quite possible to
find confirmation of this statement in the scientific data accumulated
to date. The metabolome is a molecular phenotype and is the collector
of all biochemical events in the organism since metabolites are
substrates, intermediates, and end products of biochemical reactions
combined in the metabolic pathways. Therefore, it is not surprising
that the aging of the organism has a systemic reflection in the
metabolome. Based on the previous metabolomic studies, several major
metabolic pathways, e.g., related to lipids and lipoproteins, steroid
hormones, the renal system and excretion, amino acids and muscle, diet,
oxidative stress, and inflammation, are involved in aging [[132]52]. On
the other hand, the morbidity and overall mortality from the most
common diseases with age are so obvious that it even allows us to
consider aging as a risk factor for diseases [[133]53]. These
established links between aging, metabolomes, and diseases make the
existence of connections between aging and diseases quite expected.
From this point of view, this work simply revealed this connection. The
novelty here is that evidence of this connection has been obtained with
mathematical accuracy for the ‘ome’ level, which makes it possible to
make reliable conclusions at the level of the organization of living
systems.
The revealed identity in the lists of the metabolic pathways altered in
aging and pathological conditions provokes the outline of the set of
pathways, which is essentially a metapathway (“meta-” means compiled
from several pathways). This metapathway is the main collector of both
age-related and pathological changes. The metapathway, formed by
combining six metabolic pathways, is simultaneously highly enriched
with metabolites associated with both aging and pathological conditions
([134]Table 1). The definition of this metapathway has scientific
significance because it points to the part of metabolism (amino acid
metabolism, butanoate metabolism, glyoxylate metabolism, and
dicarboxylate metabolism) associated with the development of
pathologies and aging simultaneously.
Although the study design implies that the metabolic pathways included
in the metapathway are relevant to diseases and age-related changes,
the following brief review of the accumulated scientific evidence on
the involvement of the key metabolites of these pathways in the aging
process and disease development provides additional support for the
designation of the metapathway.
* Amino Acids
Omitting the huge amount of data on the influence of amino acids on
diseases, age-related diseases, and lifespan [[135]54], below are some
facts confirming the rationale for including the four amino
acid-related pathways in the metapathway.
Branched-chain amino acids (BCAAs), including valine, leucine, and
isoleucine, are the most abundant essential amino acids, and diet is
the main significant source of BCAAs for humans [[136]55]. They play
key roles in the regulation of many physiological processes and aging.
Many studies report findings on the relationship between BCAA blood
levels and age-related changes in body composition, physical function,
sarcopenia, obesity, insulin and glucose metabolism, and the biology of
aging itself [[137]56,[138]57,[139]58,[140]59].
Studies in large population cohorts have demonstrated a
pathophysiological role for arginine metabolites in major chronic
diseases in old age, particularly for vascular disease and
atherosclerosis [[141]60]. Using a longitudinal approach,
phenylalanine, tyrosine, and tryptophan metabolism was shown to be
altered with aging in the plasma in the primate aging model
[[142]61,[143]62]. Elevated serum phenylalanine has been linked to
telomere loss in men [[144]63], inflammatory disease [[145]64], and
type 2 diabetes [[146]65]. Dietary protein restriction in rats
increased lifespan and decreased phenylalanine levels in liver
[[147]66].
Studies have been performed focusing on tryptophan in the following
aging-related disorders: cardiovascular disease
[[148]67,[149]68,[150]69], chronic kidney disease [[151]70], diabetes
[[152]71], depression [[153]72], inflammatory bowel disease [[154]73],
and multiple sclerosis [[155]74]. Tryptophan levels were most
frequently lower in the disease state, and supplemental tryptophan most
often decreased disease phenotypes. Dietary tryptophan restriction
activates specific anti-aging pathways, resulting in rats’ delayed
reproductive aging [[156]75] and extended lifespan [[157]76,[158]77].
Tyrosine is the precursor for the neurotransmitters dopamine and
norepinephrine. In a metabolomics study of human plasma, tyrosine
levels increased with age [[159]35]. Increased dietary tyrosine levels
are linked with increased cognitive performance [[160]78]. Tyrosine
supplementation enhances dopaminergic neurotransmission in Parkinson’s
disease patients [[161]79]. High levels of tyrosine in the plasma
increase the risk of type 2 diabetes [[162]65].
Increased alanine consumption may be beneficial for metabolic disorders
such as type 2 diabetes, as its addition to a cultured pancreatic beta
cell line increases both glucose metabolism and the secretion of
insulin [[163]80]. Alanine supplementation has also been shown to
stimulate the proliferation of lymphocytes [[164]81] and thymocytes
[[165]82], which may decrease the aging-related loss of immune system
function. In an animal model, dietary alanine prevents obesity
[[166]83]. Alanine levels decline with aging in mouse plasma [[167]47]
and muscle [[168]84]. The metabolism of alanine to pyruvate may play a
role in the extended longevity of nematodes [[169]85].
Aspartate participates in a range of important cellular functions, such
as the urea cycle and the malate–aspartate shuttle. Mitochondrial
electron transport chain function declines with aging [[170]86] and is
required for aspartate synthesis [[171]87]. Metals bind to aspartate to
form complexes with antioxidant function [[172]88]. Aspartate
supplementation decreases reactive oxygen species (ROS), increases ATP
levels [[173]89], and decreases hepatotoxicity and oxidative stress
[[174]90]. Aspartate regulates neurotransmission [[175]91] and may
contribute to aging-related cognitive impairment by facilitating
excitotoxicity [[176]92].
Glutamate plays an essential role in learning and cognition, but at
high levels, glutamate induces neuronal excitotoxicity, contributing to
neuronal injury in neurodegenerative disorders, including Alzheimer’s
disease [[177]93] and amyotrophic lateral sclerosis [[178]94].
Glutamate level declined in the human brain during young adulthood
[[179]95]. Glutamate binds free ammonia, which is especially toxic to
neurons [[180]96], and it serves as an important anaplerotic source of
the anti-aging citric acid cycle intermediate alpha-ketoglutarate
[[181]97]. Higher levels of glutamate were found in the plasma of
long-lived rats [[182]98].
* Butanoate Metabolism
Butanoate metabolism was defined as one of metabolic-age predictors
[[183]99]. Butanoate metabolism has been linked to sarcopenia,
deterioration of skeletal muscle, with age [[184]100]. Aging often
leads to mitochondrial dysfunction. Butyrate enhances mitochondrial
activity through serving as an additional energy source and supporting
ATP production in mitochondria [[185]101]. Additionally, butyrate
participates in the upregulation of anti-inflammatory genes and
downregulation of pro-inflammatory genes, leading to reduced systemic
inflammation associated with aging [[186]102]. This function is crucial
in preventing age-related intestinal permeability [[187]103]. Butyrate
can modulate immune responses in the gut [[188]104]. Dysregulation of
immune responses is a hallmark of aging, and butyrate’s role in immune
homeostasis is of significant interest.
Moreover, butyrate’s influence also extends to telomere shortening, a
hallmark of cellular aging. Butyrate’s impact on inflammation and
oxidative stress may indirectly influence telomere maintenance. Chronic
inflammation and oxidative stress often accelerate telomere shortening,
but by reducing these factors, butyrate may help slow down the rate of
telomere erosion [[189]105]. Additionally, butyrate’s anti-inflammatory
properties can help alleviate the pro-inflammatory secretions of
senescent cells, reducing the harmful effects associated with cellular
senescence [[190]106].
* Glyoxylate and Dicarboxylate Metabolism
Glyoxylate and dicarboxylate metabolism also have a large rise in
contribution to metabolome variance with age [[191]99]. It is strictly
linked with glycine, serine, and threonine metabolism, pyruvate
metabolism, and ascorbate metabolism, all of which have been identified
as linked to aging [[192]99]. This pathway has already been found to be
associated with many metabolic diseases (type 2 diabetes, obesity, and
atherosclerosis) [[193]107,[194]108] and microbiome composition
[[195]109], changes in which can contribute to metabolic inflammation
and major age-related metabolic disorders [[196]110,[197]111,[198]112].
Thus, accumulated scientific evidence points to the relevance of the
pathways comprising the metapathway to aging and disease development.
It is necessary to explain why the aminoacyl-tRNA biosynthesis pathway
was not included in the metapathway. In this pathway, amino acids are
linked with their cognate transfer RNAs. The metapathway also largely
consists of amino acid-related pathways. Therefore, double accounting
of many amino acids would violate the statistical model of MSEA. The
feasibility of including this pathway in the metapathway after leveling
the double hit of amino acids remains to be studied.
As it was explained earlier, several steroids from the aging-related
studies were not recognized by MetaboAnalyst, which may have prevented
the steroid hormone biosynthesis pathway from being included in the
metapathway. However, the association of steroids with aging is well
established, and this pathway ranks first among metabolic pathways in
correlation and prediction of chronological age [[199]113]. Thus, there
is a reason to investigate the rationale for including this metabolic
pathway in the metapathway in the future.
The next remark concerns the fact that metabolic pathways have a
holistic structure. The integrity of the metapathway is also present.
The pathways that comprise the metapathway are locally located in the
global pathway network ([200]Supplementary Figure S3) and are
connected, except for the phenylalanine, tyrosine, and tryptophan
biosynthesis pathways. The metapathway was defined from the
experimental data, where gaps are not uncommon. However, it is
conceivable that this gap could be filled by the glycine, serine, and
threonine metabolism or the TCA cycle. The first one was only
associated with pathological conditions, but the relationship of
related amino acids with aging is well known from the literature
[[201]54]. The TCA cycle presented only in age-related pathways is the
central metabolic pathway and may be quite suitable for obtaining
metapathway integrity.
The construction of a metabolite–metabolite interaction network
confirmed the high functional integrity of the metapathway ([202]Figure
3, [203]Table 2), indicating the central role of energy metabolism in
it. This is close to the mitochondrial theory of aging, where the
energy metabolism is disrupted due to mitochondrial dysfunction. There
is room for theorizing here. Since the metapathway is a common
collector of metabolic changes during aging and various pathologies,
with energy metabolism playing a central role in it, aging associated
with disturbances in energy metabolism leads to deterioration in
health. Conversely, healthy aging is characterized by the absence of
significant changes, or only moderate changes, in energy metabolism.
Among the prospects for practical application, the use of the
metapathway in metabolomic studies of aging and age-related diseases
stands out. In metabolomics, enrichment analysis is extremely common.
MetaboAnalyst, which was used in this work to perform enrichment
analysis, has been used by >500,000 researchers
([204]https://www.metaboanalyst.ca/docs/UserStats.xhtml; accessed on
September 4, 2024). It allows to combine pathways and perform MSEA to
return a p-value for the enrichment of the metapathway by altered
metabolites. Such p-value is closely related to metapathway alteration
and can be used in the study of human health in association with aging.
The measure of metapathway alteration is also a potential way to create
health-related metabolic clocks and can be useful in the study of
healthy aging.
5. Conclusions
It was found that aging and pathological conditions at the metabolic
level are part of the same processes, with a probability of 99.96%.
This fact gives a new perspective on the positioning of aging and
disease development as different ways of looking at the same
phenomenon. The obtained high probability also provided the basis for
identifying a metapathway that is simultaneously related to both aging
and disease development, which is of fundamental importance. Based on
metapathway composition, it was hypothesized that healthy aging can be
considered as aging without alteration of energy metabolism. The
practical application of the metapathway has enormous potential.
Analysis of metapathway enrichment by the MSEA, which is widely used in
metabolomics, may provide a metric of age and health-related changes in
the organism, which is essentially the basis for developing
health-related metabolic clocks.
Supplementary Materials
The following supporting information can be downloaded at:
[205]https://www.mdpi.com/article/10.3390/metabo14110593/s1, Figure S1:
Frequency of occurrence for metabolites in untargeted metabolomics
studies related to aging; Figure S2: Frequency of detection of abnormal
metabolite concentration in different human body conditions; Figure S3:
Projection of metapathway metabolites on the KEGG global pathway
network; Table S1: List of untargeted metabolomics studies on aging and
identified aging-associated metabolites; Table S2: List of metabolites
with abnormal concentrations and associated conditions; Table S3:
Metabolite–metabolite interaction network parameters for nodes
(metabolites).
[206]metabolites-14-00593-s001.zip^ (1.1MB, zip)
Author Contributions
Conceptualization, methodology, investigation, formal analysis, data
curation, and writing—original draft preparation, P.G.L. and E.E.B.;
writing—review and editing, D.L.M., O.P.T. and A.I.A.; finding
acquisition, A.I.A. All authors have read and agreed to the published
version of the manuscript.
Institutional Review Board Statement
Not applicable.
Informed Consent Statement
Not applicable.
Data Availability Statement
The data are presented in [207]Supplementary Materials.
Conflicts of Interest
The authors declare no conflicts of interest.
Funding Statement
This work was financed by the Ministry of Science and Higher Education
of the Russian Federation within the framework of Agreement No.
075-15-2024-643.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data
contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI
and/or the editor(s) disclaim responsibility for any injury to people
or property resulting from any ideas, methods, instructions or products
referred to in the content.
References