TYPE Original Research

PUBLISHED 14 September 2023

DOI 10.3389/fmolb.2023.1145279

The p53 tumor suppressor

OPEN ACCESS regulates AKR1B1 expression, a
EDITED BY
José Díaz-Chávez,
Instituto Nacional de Cancerología
metastasis-promoting gene in
(INCAN), Mexico

REVIEWED BY
breast cancer
Wangxiao He,
Xi’an Jiaotong University, China
Harish Chander,
Carolina Di Benedetto 1, Carla Borini Etichetti 2,
National Institute of Biologicals, India Nabila Cocordano 3, Alejo Cantoia 4, Evelyn Arel Zalazar 3,
*CORRESPONDENCE Silvio Bicciato 5, Mauricio Menacho-Márquez 3,
Javier Girardini,
girardini@idicer-conicet.gob.ar Germán Leandro Rosano 4 and Javier Girardini 3*
RECEIVED 15 January 2023 1
Department of Radiation Oncology, University of California San Francisco, San Francisco, CA,
ACCEPTED 28 August 2023 United States, 2Instituto de Fisiología Experimental de Rosario (IFISE), Consejo Nacional de
PUBLISHED 14 September 2023 Investigaciones Científicas y Técnicas (CONICET), Universidad Nacional de Rosario, Rosario, Argentina,
3
CITATION
Instituto de Inmunología Clínica y Experimental de Rosario (IDICER), Consejo Nacional de
Di Benedetto C, Borini Etichetti C, Investigaciones Científicas y Técnicas (CONICET), Universidad Nacional de Rosario, Rosario, Argentina,
4
Cocordano N, Cantoia A, Arel Zalazar E, Unidad de Espectrometría de Masa, Instituto de Biología Molecular y Celular de Rosario (IBR), Consejo
Bicciato S, Menacho-Márquez M, Nacional de Investigaciones Científicas y Técnicas (CONICET), Universidad Nacional de Rosario, Rosario,
Rosano GL and Girardini J (2023), The Argentina, 5Department of Life Sciences, University of Modena and Reggio Emilia, Modena, Italy
p53 tumor suppressor regulates
AKR1B1 expression, a metastasis-
promoting gene in breast cancer.
Front. Mol. Biosci. 10:1145279.
doi: 10.3389/fmolb.2023.1145279 Alteration of metabolism in cancer cells is a central aspect of the mechanisms that
sustain aggressive traits. Aldo–keto reductase 1 B1 (AKR1B1) catalyzes the
COPYRIGHT
© 2023 Di Benedetto, Borini Etichetti, reduction of several aldehydes to alcohols consuming NADPH. Nevertheless,
Cocordano, Cantoia, Arel Zalazar, the ability of AKR1B1 to reduce different substrates renders difficult to
Bicciato, Menacho-Márquez, Rosano and
comprehensively ascertain its biological role. Recent evidence has implicated
Girardini. This is an open-access article
distributed under the terms of the AKR1B1 in cancer; however, the mechanisms underlying its pro-oncogenic
Creative Commons Attribution License function remain largely unknown. In this work, we report that
(CC BY). The use, distribution or
AKR1B1 expression is controlled by the p53 tumor suppressor. We found that
reproduction in other forums is
permitted, provided the original author(s) breast cancer patients bearing wild-type TP53 have reduced AKR1B1 expression.
and the copyright owner(s) are credited In cancer cell lines, p53 reduced AKR1B1 mRNA and protein levels and repressed
and that the original publication in this
promoter activity in luciferase assays. Furthermore, chromatin
journal is cited, in accordance with
accepted academic practice. No use, immunoprecipitation assays indicated that p53 is recruited to the
distribution or reproduction is permitted AKR1B1 promoter. We also observed that AKR1B1 overexpression promoted
which does not comply with these terms.
metastasis in the 4T1 orthotopic model of triple-negative breast cancer.
Proteomic analysis of 4T1 cells overexpressing AKR1B1 showed that
AKR1B1 exerts a marked effect on proteins related to metabolism, with a
particular impact on mitochondrial function. This work provides novel insights
on the link between the p53 pathway and metabolism in cancer cells and
contributes to characterizing the alterations associated to the pathologic role
of AKR1B1.

KEYWORDS

aldo–keto reductase 1 B1, p53, metastasis, breast cancer, aldo–keto reductases, electron
transport chain, mitochondria, proteomics

Frontiers in Molecular Biosciences 01 frontiersin.org

Di Benedetto et al. 10.3389/fmolb.2023.1145279

1 Introduction downregulation of genes of the mevalonate pathway (Moon et al.,

2019; Borini Etichetti et al., 2020).
Aldo–keto reductase 1 B1 (AKR1B1), a cytosolic enzyme Considering that p53 exerts a concerted regulation of different
belonging to the aldo–keto reductase superfamily, catalyzes the aspects of cell metabolism, we hypothesized that it might also
reduction of several aldehydes to alcohols consuming NADPH affect enzymes involved in the polyol pathway. Indeed, our work
(Pastel et al., 2012). Glucose reduction by AKR1B1 has attracted shows that AKR1B1 expression is repressed at the transcriptional
attention because of its consequences in diabetes (Tang et al., 2012). level by p53 in cancer cell lines and that AKR1B1 mRNA levels are
This is the first step of the polyol pathway, which consists of two reduced in breast cancer patients’ tumors retaining wild-type
reactions that transform glucose into fructose. The first reaction is TP53 (the gene encoding p53), as compared to tumors bearing
the reduction of glucose to sorbitol catalyzed by AKR1B1, the rate- mutations in this gene. Using an immunocompetent mouse model
limiting step of the pathway (Kinoshita, 1990). The pathway is of breast cancer, we also found that AKR1B1 overexpression
completed by the oxidation of sorbitol to fructose, catalyzed by accelerates tumor growth and promotes metastasis. Thus, our
sorbitol dehydrogenase, using NAD+ as a cofactor. Fructose may be results indicate that loss of the wild-type (wt) p53 repressive effect
later phosphorylated to fructose-6-P and reincorporated into on AKR1B1 expression could collaborate with breast cancer
glycolysis or used to generate advanced glycation products aggressiveness. In addition, proteomic analyses performed in
(Srivastava et al., 2005). Recent experimental evidence has shown breast cancer cells to explore the mechanisms underlying the
that AKR1B1 may exert pro-oncogenic effects. For instance, in AKR1B1 pro-oncogenic role indicate that AKR1B1 overexpression
breast cancer cell lines, AKR1B1 was shown to cooperate with highly impacts cell metabolism.
migration and invasion in vitro (Wu et al., 2017b). Likewise,
AKR1B1 knockdown reduced tumor formation in vivo and lung
colonization upon vein tail injection in immunocompromised mice. 2 Results
In addition, levels of cancer stem cell markers were reduced upon
AKR1B1 silencing in lung and breast cancer cell lines (Schwab et al., To investigate whether the p53 pathway affects AKR1B1
2018). To date, a comprehensive understanding of the biological role expression, we first analyzed public cancer patients’ databases
of AKR1B1 remains elusive, mostly due to its ability to reduce other looking for correlations between p53 status and AKR1B1 mRNA
substrates. These include aldoses such as xylose and glyceraldehyde, levels. We focused on breast cancer patients from METABRIC and
as well as methylglyoxal, prostaglandin H2 (PGH2), and lipid TCGA databases, both of which contain information on the
peroxides, among others (Bresson et al., 2012; Tammali et al., mutational status of the TP53 gene. We found that patients
2012). Therefore, the available evidence suggests that alteration of retaining wt TP53 alleles showed a significant reduction in
different aspects of cell metabolism underlies the pathologic effects AKR1B1 mRNA levels compared with patients with mutations in
of AKR1B1. this gene (Figures 1A,B). These results indicate that inactivation of
Alteration of metabolism is a hallmark of cancer cells (Hanahan TP53 is correlated with enhanced AKR1B1 mRNA levels in breast
and Weinberg, 2011). Following the initial observations showing a cancer patients and suggested us that p53 negatively regulates
switch to aerobic glycolysis and lactate production, intense research AKR1B1 expression. To further explore this hypothesis, we
in the last years has brought a more complex scenario to light. wondered whether p53 may affect AKR1B1 expression in cancer
Cancer cells often show enhanced nutrient uptake and extensive cell lines. First, we performed Western blot analysis in a panel
reshaping of intermediate metabolite usage among anabolic and including human and murine cell lines (Figure 2A). Interestingly, we
catabolic pathways (Pavlova and Thompson, 2016). A growing body observed striking differences in the expression levels of this enzyme,
of evidence has documented the role of the tumor suppressor p53 in which was present in some cell lines but not detected in others. In
the regulation of cell metabolism. p53 constitutes the central hub of a line with this observation, AKR1B1 mRNA expression levels
complex signaling pathway activated in response to different stress obtained from RNA sequencing data publicly available on the
signals (Kastenhuber and Lowe, 2017). Upon stabilization and Cancer Cell Line Encyclopedia showed a similar expression
activation, p53 can regulate the transcription of specific target pattern (Supplementary Figure S1A). In addition,
genes depending on the stimulus and context. AKR1B1 mRNA and protein expression levels are highly
The p53 pathway can control different aspects of glucose correlated in breast cancer cell lines (Supplementary Figure S1B),
metabolism. For example, p53 represses the expression of glucose suggesting that AKR1B1 protein levels are mainly regulated at the
transporters (Schwartzenberg-Bar-Yoseph et al., 2004) and transcriptional level. It should be noted that, considering the human
negatively regulates glycolysis through its transcriptional targets breast cell lines on Figure 2A, AKR1B1 protein expression seems to
TIGAR (Bensaad et al., 2006), mir-34a (Napoli and Flores, 2017), correlate with p53 status, since the two cell lines that express high
and Parkin (Zhang et al., 2011). p53 also enhances the tricarboxylic AKR1B1 protein levels (MDAMB231 and MDAMB468) have
acid (TCA) cycle by repressing lactate release on the extracellular mutations in TP53, while AKR1B1 protein expression was not
medium and inducing GSL2 expression (Napoli and Flores, 2017). detected in wt p53 cells (MCF10A and MCF7).
The emerging picture suggests that p53 contributes to maintaining To investigate whether p53 represses AKR1B1 expression, we
moderate levels of glucose uptake and glycolytic flux, in concert with concentrated on HEK-293 and HepG2 cell lines, in which p53 is not
tightly regulated oxidative phosphorylation. In addition, mutated and AKR1B1 was detectable by Western blot. We found
p53 inhibits the activation of SREBP1/2 transcription factors, that AKR1B1 mRNA levels increased upon silencing of endogenous
leading to different effects on lipid metabolism, including the p53, and a similar effect was observed on AKR1B1 protein levels

Frontiers in Molecular Biosciences 02 frontiersin.org

Di Benedetto et al. 10.3389/fmolb.2023.1145279

FIGURE 1
AKR1B1 RNA expression is linked to TP53 status in breast cancer. Standardized RNA levels of AKR1B1 RNA are lower in wild-type TP53 tumors than in
mutant TP53 tumors in breast cancer samples from (A) METABRIC (n = 1866) and (B) TCGA (n = 994). The two-tailed Mann–Whitney test was performed;
****p < 0.0001.

FIGURE 2
p53 represses AKR1B1 expression. (A) AKR1B1 protein levels are highly variable in a panel of human and murine cell lines, as observed by Western blot
analysis. Colored boxes represent TP53 status, obtained from the Cancer Cell Line Encyclopedia or from the literature (Yerlikaya and Erin, 2008). (B)
Silencing of p53 increases AKR1B1 expression in (B) HEK-293 and (C) HepG2 cells. Cells were transfected with siRNA-targeting p53 or control (sip53 or
siControl), and AKR1B1 expression levels were determined by qPCR or Western blot. p53 silencing was confirmed by Western blot. Intensity of
Western blot bands was quantified using ImageJ and normalized to β-actin levels. Two-tailed t-test, *p < 0.05, **p < 0.01, and ***p < 0.001; n = 4.

(Figures 2B,C), suggesting that wt p53 represses AKR1B1 expression transfection with pAKR1B1luc in H1299 cells (p53 null), we
at the transcriptional level in these cancer cell lines. found that p53 inhibited promoter activity in luciferase assays
To further characterize the role of p53 on AKR1B1 expression, (Figure 3A). To identify sequences involved in the repressive
we analyzed its effect on promoter activity. We generated the effect of p53 on the AKR1B1 promoter, we generated a series of
reporter pAKR1B1luc by cloning a fragment of the AKR1B1 reporters containing deletions at the 5′end of the promoter, covering
promoter, including the region −1785 to +24 from the the region from −1785 to −14 (Supplementary Figure S2A). When
transcription start site, into the pGL3 vector. Upon co- we tested these reporters in luciferase assays, we found that

Frontiers in Molecular Biosciences 03 frontiersin.org

Di Benedetto et al. 10.3389/fmolb.2023.1145279

FIGURE 3
p53 is recruited to the AKR1B1 promoter. (A) p53 reduces AKR1B1 promoter activity. Luciferase assays were performed in H1299 cells co-transfected
with the indicated reporter plasmid for the AKR1B1 promoter deletions (pAKR1B1luc; pAKR1B1luc100; pAKR1B1luc70; or pAKR1B1luc40), a plasmid
expressing p53 (pCDNA3-p53) or empty vector (pCDNA3) as control, and pCMV-β-gal. Values were normalized to β-galactosidase activity and expressed
as fold change relative to cells transfected with empty vector. Two-way ANOVA test, **p < 0.01, ***p < 0.001, and ****p < 0.0001; n = 4 for
pAKR1B1luc; and n = 3 for pAKR1B1luc100, pAKR1B1luc70, and pAKR1B1luc40. (B) p53 is recruited to the AKR1B1 promoter in H1299 (p53 null) cells
transfected with pCDNA3-p53 or pCDNA3 as control. ChIP immunoprecipitation was performed using the anti-p53 antibody (DO1). (C) Endogenous
p53 is recruited to the AKR1B1 promoter in HEK-293 cells. ChIP immunoprecipitation was performed using the anti-p53 antibody (DO1) or a control
antibody. (D) Endogenous p53 is recruited to the AKR1B1 promoter in HepG2 cells. ChIP immunoprecipitation was performed using the anti-p53 antibody
(DO1) or a control antibody. A two-tailed t-test was performed, *p < 0.05 and **p < 0.01.

FIGURE 4
AKR1B1 promotes invasion in vitro and metastasis in vivo. (A) Kaplan–Meier plot showing breast cancer patients’ overall survival probability over time
based on AKR1B1 mRNA expression. The analysis was performed using RNA-Seq data available on KM plotter for 2,976 breast cancer patients. Red and
black curves represent samples that show high and low gene expression levels, respectively, relative to the median AKR1B1 expression value. High levels
of AKR1B1 RNA are significantly correlated with breast cancer patients’ poor prognosis. A log-rank test was performed. (B) Western blot analysis
confirms AKR1B1-GFP overexpression in 4T1 cells. (C) AKR1B1 overexpression in 4T1 cells increases aldose reductase activity. Enzymatic activity was
quantified in cell extracts, using D + -xylose as the substrate and following absorbance at 340 nm, corresponding to NADPH oxidation. A two-tailed t-test
was performed, *p < 0.05. (D) AKR1B1 overexpression in 4T1 cells increases invasion in vitro. 4T1-AKR1B1 (AKR1B1) or 4T1-GFP (GFP) cells were plated on
the upper side of Matrigel-coated Transwell chambers. After 24 h, cells on the lower surface were stained and quantified. Representative images are
shown. A two-tailed t-test was performed, *p < 0.05; n = 4. (E) 4T1-AKR1B1 or 4T1-GFP cells were injected in the fat pad of female BALBc mice, and tumor
growth was monitored (4T1-AKR1B1 n = 7, 4T1-GFP n = 10). Non-linear regression for exponential growth fit was performed, ****p < 0.0001. (F)
AKR1B1 overexpression increases metastatic potential of 4T1 cells to the lung. After sacrifice, lung metastases were analyzed by intra-tracheal injection of
India ink. Pictures show representative lungs obtained from the two groups. A two-tailed t-test was performed, ****p < 0.0001.

Frontiers in Molecular Biosciences 04 frontiersin.org

Di Benedetto et al. 10.3389/fmolb.2023.1145279

p53 repressed AKR1B1 promoter activity in all cases (Figure 3A), tumor development was monitored periodically. As seen in
indicating that DNA regions that sustain the observed effect of Figure 4E, tumors arising from 4T1-AKR1B1 cells showed a
p53 are located close to the transcription start site on the AKR1B1 significant increase in the growth rate, and the final mean tumor
promoter. volume approximately doubled the mean volume of the control
Next, we performed chromatin immunoprecipitation (ChIP) condition (1309.4 mm3 versus 673.4 mm3, respectively). Of note,
assays to analyze if p53 is recruited to the AKR1B1 promoter. Upon three mice in the AKR1B1-overexpressing group died before
expression of p53 in H1299 cells and immunoprecipitation with an experiment completion, at days 6, 35, and 37 post-injection. We
anti-p53 antibody, we found a significant enrichment of the region also analyzed the presence of macroscopic metastases in the lungs.
between −46 and +24 of the AKR1B1 promoter in the We observed that all mice in our study developed lung metastasis.
immunoprecipitated DNA, compared to H1299 control cells However, the overexpression of AKR1B1 resulted in a dramatic
transfected with an empty vector and, therefore, not expressing increase in metastasis development, as mice injected with 4T1-
p53 (Figure 3B). We performed similar experiments but AKR1B1 showed 37.43 pulmonary metastases on average,
immunoprecipitating endogenous p53 in HEK-293 and whereas 4T1-GFP control mice showed an average of five
HepG2 cell lines. Here, a significant enrichment of the AKR1B1 pulmonary metastases (Figure 4F), i.e., 4T1-AKR1B1 tumors
promoter fragment was found relative to the control condition, in produced a >7-fold increase in the number of lung metastases
which DNA was immunoprecipitated with an unrelated antibody compared to the control. As it can be observed in the
(Figures 3C,D). Moreover, we also found an AKR1B1 promoter representative images shown in Figure 4F, pulmonary metastases
enrichment when we performed ChIP experiments upon were also much bigger in the 4T1-AKR1B1 group than in the control
p53 protein stabilization upon doxorubicin treatment in mice. Likewise, three metastases were detected in the spleen and one
HCT116 cells (Supplementary Figure S2B). Taken together, our in the liver in the group of mice injected with 4T1-AKR1B1 cells, but
data show that p53 represses AKR1B1 transcription and that it is macroscopic metastases were not detected in these organs in the
recruited to the AKR1B1 promoter. In silico analysis of the AKR1B1 control group (results not shown). Together, these results show, for
promoter sequence failed to identify a p53 responsive element, the first time, that AKR1B1 overexpression promotes metastasis in
suggesting that p53 is recruited by an indirect mechanism, an orthotopic breast cancer model in vivo.
probably involving the interaction with other proteins present on To gain more insight into the mechanisms underlying the
the chromatin. oncogenic role of AKR1B1, we performed a global proteomic
Our results suggest that AKR1B1 levels are tightly regulated in analysis. Protein extracts from 4T1-AKR1B1 and 4T1-GFP were
the presence of active p53. In contrast, p53 inactivation may analyzed following a label-free bottom–up mass spectrometry
promote AKR1B1 overexpression. Therefore, we hypothesized strategy. Based on our analysis, 3822 proteins were identified
that AKR1B1 overexpression may cooperate with tumor with at least two unique peptides and quantified. As evidenced in
progression. Accordingly, when we analyzed public databases, we the volcano plot in Figure 5A, 99 proteins showed significantly
found that breast cancer patients with high AK1B1 mRNA levels altered abundances with a fold change (FC) of 2 or higher. Among
displayed a significant reduction in overall survival, with a survival them, 62 proteins were upregulated and 37 downregulated in 4T1-
probability of 0.807 and 0.872 for patients expressing high and low AKR1B1 cells compared with 4T1-GFP cells. Accordingly, we
AKR1B1 mRNA, respectively (n = 2976, Figure 4A). Nevertheless, defined a list of differentially expressed proteins (DEPs), up- or
the role of AKR1B1 overexpression on metastasis has not been downregulated, considering only proteins identified by two or more
studied. Therefore, we took advantage of an orthotopic model of unique peptides with an FC of 2 or higher (p-value ≤ 0.05,
breast cancer, consisting in the transplantation of 4T1 cells in the Supplementary Table S1).
mammary fat pad of BALB/c mice. These cells are derived from a We performed a functional analysis of the identified DEPs to
murine triple-negative breast adenocarcinoma and develop tumors explore the effects of AKR1B1 overexpression on biological
in situ, which can metastasize to other tissues, primarily the lungs processes. Using the KEGG Mapper tool, we performed a
(Lelekakis et al., 1999). In contrast to direct injection of tumor cells pathway enrichment analysis. We found that the category
in the vein tail, this approach has the advantage of providing a model metabolic pathways was the most enriched in the DEPs
for the complete metastatic process. Moreover, the transplantation identified by our proteomic study. As shown in Figure 5B,
of murine cells allows the use of immunocompetent mice, thus this category includes 15 DEPs, compared with all the other
considering the role of the immune system in the analysis (Aslakson categories that counted nine DEPs or less. These results
and Miller, 1992). 4T1 cells with stable overexpression of GFP- reinforce the idea that AKR1B1 contributes to shape tumor
AKR1B1 (4T1-AKR1B1) or GFP as control (4T1-GFP) were cell metabolism. Among the upregulated proteins in the
generated, and the presence of the fusion protein was confirmed metabolic pathways category, we found several components
by Western blot (Figure 4B). We also observed an increase in aldose of the mitochondrial electron transport chain (Figure 5C).
reductase activity in extracts from 4T1-AKR1B1 cells, corroborating For example, components of the ubiquinone oxidoreductase
the overexpression of a functional protein (Figure 4C). complexes Ndufs5, Ndufv2, Ndufa5, Ndufa8, Ndufc2, and
Since the effect of AKR1B1 overexpression on cell invasion Ndufaf2 were identified. Similarly, Coq3, a protein involved
in vitro has not been analyzed previously, we first performed in coenzyme Q biosynthesis, was found upregulated. These
Transwell assays on Matrigel-coated filters. We found that 4T1- results suggest that AKR1B1 overexpression may enhance
AKR1B1 cells showed a significant increase in invasive behavior oxidative phosphorylation. In agreement, previous reports
(Figure 4D). Next, 4T1-AKR1B1 or AKR1B1-GFP cells were showed that AKR1B1 downregulation reduced the oxygen
injected into the mammary fat pad of female BALB/c mice, and consumption rate in the A549 cell line (Schwab et al., 2018).

Frontiers in Molecular Biosciences 05 frontiersin.org

Di Benedetto et al. 10.3389/fmolb.2023.1145279

FIGURE 5
AKR1B1 overexpression highly impacts metabolism in 4T1 cells. (A) Volcano plot showing differential protein expression in 4T1-AKR1B1 cells relative
to 4T1-GFP cells. Proteins identified by at least two unique peptides are represented (3822). Only proteins identified by at least two peptides, showing FC ≥
2 (up- or downregulated) and p-value ≤ 0.05, were considered as differentially expressed proteins. Upregulated proteins are shown as red dots and
downregulated as blue dots. The gray dots represent the proteins with unaltered expression. (B) Pathway enrichment analysis by the KEGG Mapper.
Bar graph showing the KEGG categories enriched in DEPs from the proteomic analysis. The category “Metabolic pathways” included 15 DEPs,
representing 15% of the DEPs identified. (C) Functional KEGG categories enriched in proteins affected by AKR1B1 overexpression. Heatmap showing
selected functional categories defined using the KEGG mapper tool containing more than four DEPs. The color code indicates relative fold expression
expressed as log2 FC, comparing 4T1-AKR1B1 with 4T1 GFP cells. The numbers correspond to the log2 FC of each DEP. (D) ROS levels decrease in 4T1-
AK1B1 cells. 4T1-AKR1B1 or control cells were incubated with 5 µM 2′,7′-dichlorofluorescein diacetate for 45 min, and after washing, fluorescence was
measured. Non-stained cell suspensions were used as blank. A two-tailed t-test was performed, ***p < 0.001. (E) A cluster related to the mitochondrial
electron transport chain was defined by network analysis. The color code indicates relative fold expression expressed as log2 FC. Hexagons represent
components of the mitochondrial electron transport chain. The solid black lines indicate functional and/or physical interaction defined by the software.

To further characterize the effect of AKR1B1 on oxidative stress, et al., 2011; Sánchez-Gómez et al., 2016), suggesting that its
we measured reactive oxygen species (ROS) levels in 4T1- detoxifying activities during the antioxidant response could
AKR1B1 cells. We found that AKR1B1 overexpression counterbalance the consequences of increased oxidative
significantly reduced ROS levels compared to control cells phosphorylation.
(Figure 5D). AKR1B1 is known to protect cells against We used the STRING protein interaction repository to outline
reactive compounds generated under oxidative stress (Shen the interaction networks between the identified DEPs that could

Frontiers in Molecular Biosciences 06 frontiersin.org

Di Benedetto et al. 10.3389/fmolb.2023.1145279

help understand their functional relationships. In this case, we exerts a marked effect on proteins related to metabolism in
extended the DEP list considering proteins with log2FC ≥ 0.85 or breast cancer cells. We also showed that p53 represses
log2FC ≤ −0.85 (p-value ≤ 0.05), in order to increase the number of AKR1B1 expression by inhibiting its promoter activity. The
interactors and, therefore, to highlight connections that may be analysis of breast cancer patients’ databases showed a significant
unappreciated with a more restrictive criterion (Supplementary decrease in AKR1B1 RNA expression in tumors retaining wt
Table S1). The interactions retrieved from STRING were loaded TP53. Based on our results, we hypothesize that p53 contributes
onto the Cytoscape platform to perform network analysis, to tightly regulating AKR1B1 transcript levels in physiological
visualization, and clustering. As shown in Figure 5E, one highly conditions. During tumor progression, blockade of
interconnected cluster emerged from the network obtained p53 function may contribute to increase AKR1B1 RNA levels.
(Supplementary Figure S3). This cluster contains upregulated Mutations in TP53 are among the most frequent alterations in
proteins related to mitochondrial function, including the human cancer. In breast cancer, TP53 mutation is found in
components of the ubiquinone oxidoreductase complex also approximately 30% of cases, and it is correlated with poor
identified by pathway enrichment analysis in KEGG, as well as clinical outcome (Silwal-Pandit et al., 2014). In cases
Cox6c and Park7. Cox6c is a subunit of cytochrome c oxidase, maintaining wt alleles, p53 function is often inhibited by the
another key component of the electron transport chain. Although aberrant activity of regulators (Hainaut and Pfeifer, 2016).
Park7 is not included in the same functional category by the KEGG AKR1B1 overexpression in cancer may also be a consequence
mapper, it is related to mitochondria. Park7 loss is associated to of additional p53-independent mechanisms. For instance,
mitochondria fragmentation and reduced membrane potential AKR1B1 transcription is enhanced by Twist2, a well
(Wang et al., 2016). Interestingly, this protein has been shown to characterized promoter of epithelial–mesenchymal transition
promote oncogenic mechanisms in breast cancer (Kim et al., 2021). that cooperates with tumor aggressiveness (Wu et al., 2017b).
Other DEPs present in the KEGG metabolic pathways Our work provides the first demonstration that
category on Figure 5C are also related to different aspects of AKR1B1 overexpression enhances metastasis from primary
metabolism, such as protein glycosylation and amino acid and tumors, in a model of breast cancer in immunocompetent
nucleotide biosynthesis. Nevertheless, a clear functional mice. These results provide strong support to the notion that
interaction among them is not so evident, and a strong an increase on AKR1B1 levels during tumor progression may
network could not be clearly defined in our network analysis. constitute a critical event to develop an aggressive phenotype.
In addition, some DEPs related to lipid metabolism were also The analysis of the effect of AKR1B1 overexpression on the
present in the metabolic pathways category. For example, proteome demonstrated a marked impact on proteins related to
Echs1 was upregulated upon AKR1B1 overexpression. This metabolism. Our results suggest that the enhancement of the
enzyme catalyzes the hydration of short-chain enoyl-CoAs, mitochondrial electron transport chain could be linked to
as part of the process of fatty acid β-oxidation in AKR1B1 oncogenic function. Moreover, we found that
mitochondria (Hu et al., 2022). Among downregulated AKR1B1 overexpression reduced ROS levels in 4T1 cells.
proteins, we found Cbr4, a mitochondrial carbonyl reductase Although clearly involved in tumorigenesis and tumor
involved in mitochondrial fatty acid synthesis (FAS) progression, accumulating evidence has made it clear that the
(Venkatesan et al., 2014). We also found Pnpla2 as role of ROS in cancer cell behavior is extremely complex and
downregulated. Pnpla2 catalyzes the first step of lipolysis, context dependent (Cheung and Vousden, 2022). It can be
generating diacylglycerols from triglycerides. Although its hypothesized that reduction of ROS levels by AKR1B1 could
role in cancer is not fully understood, some evidence protect cells from excessive damage under conditions of
indicates a cooperation with tumor suppression mechanisms. enhanced electron transport chain activity, as suggested by our
For example, in humans, PNPLA2 deletion was found in proteomic studies.
sarcoma and liposarcoma, and a Pnpla2/Hsl double-knockout We also observed an alteration of mitochondrial enzymes
mice model developed liposarcoma (Wu et al., 2017a). Taken related to fatty acid metabolism that suggests an enhancement of
together, our proteomic analyses show that fatty acid β-oxidation. The alteration of lipid metabolism in cancer
AKR1B1 overexpression deeply alters cell metabolism. cells often includes enhanced fatty acid β-oxidation as a source of
NADH, FADH2, and NADPH (Hu et al., 2022), while relying in
cytosolic FAS for lipid generation (Röhrig and Schulze, 2016). In this
3 Discussion context, Cbr4 downregulation may reduce acetyl-CoA consumption
for mitochondrial FAS, favoring its bioavailability for other
The acquisition of aggressive traits in cancer cells is often reactions. Collectively, these results suggest that the alteration of
associated to extensive reshaping of metabolic circuits. Such mitochondrial function is a central feature of AKR1B1 pro-
alterations may create a dependence on specific metabolic oncogenic function.
programs, thus providing potential avenues for therapeutic In summary, our work identifies AKR1B1 overexpression as an
intervention. However, in order to successfully interfere with alteration related to loss of p53 function that promotes the
cancer-specific metabolic features, a more profound knowledge acquisition of aggressive tumor phenotypes. Our proteomic
of the mechanisms that regulate cancer cell metabolism is analysis contributes to understanding the effects associated to
required. AKR1B1 pathologic function, showing a complex scenario where
In this work, we showed that aldo–keto reductase AKR1B1 is AKR1B1 affects different aspects of cell physiology with a marked
a strong promoter of metastasis in vivo, and its upregulation impact on metabolism.

Frontiers in Molecular Biosciences 07 frontiersin.org

Di Benedetto et al. 10.3389/fmolb.2023.1145279

4 Materials and methods PCR on genomic DNA from MDAMB231 cells and cloned into the
pGL3-basic vector using MluI/XhoI sites. PCR primers used were
4.1 Analyses of cancer datasets AKR1B1luc_F: AAAACGCGTCAGAGGCAATGGGGGATGTTA,
AKR1B1luc R:AAACTCGAGGGCGCGTACCTTTAAATAGCC.
AKR1B1 mRNA expression (as z-scores relative to all samples) pAKR1B1luc was used as a PCR template to generate a series of
and TP53 status data from The Cancer Genome Atlas (TCGA) 5′-terminally truncated AKR1B1 promoter fragments:
breast cancer dataset (Koboldt et al., 2012) or the Molecular pAKR1B1luc100, pAKR1B1luc70, and pAKR1B1luc40. Each
Taxonomy of Breast Cancer International Consortium fragment was ligated between SacI/HindIII sites into pGL3 basic
(METABRIC) study (Curtis et al., 2012) were downloaded from vector to generate the corresponding reporter. PCR primers used to
cBioPortal (Cerami et al., 2012). AKR1B1 RNA expression levels in generate these constructs were AKR1B1luc100_F: AAAGAGCTC
wt and mutant TP53 tumors were compared using the CGCAACCAATCAGAAGGCTCC, AKR1B1luc70_F: AAAGAG
Mann–Whitney test. Kaplan–Meier (KM) plots for CTCCTCCTTCGCGCAGCGGC, AKR1B1luc40_F: AAAGAG
AKR1B1 RNA expression in breast cancer were performed using CTCTTTCTGCCGACCTCACGG, AKR1B1lucHind_R: AAA
KM plotter (https://kmplot.com/analysis/) (Győrffy, 2021). Patients AAGCTTGGCGCGTACCTTTAAATAGCC
were divided into two groups based on the median
AKR1B1 expression value, and KM curves were compared using
the log-rank (Mantel–Cox) test. AKR1B1 mRNA expression (as 4.4 Cell transfection and retroviral
reads per kilobase per million mapped reads, RPKM), transduction
AKR1B1 protein levels, and TP53 status data in cancer cell lines,
publicly available on the Cancer Cell Line Encyclopedia (Ghandi DNA and siRNA (10 pmol/cm2) transfection were performed
et al., 2019; Nusinow et al., 2020), were downloaded from cBioPortal. with Lipofectamine 2000 (Invitrogen) according to the
manufacturer’s instructions. The following siRNA sequences were
used (Girardini et al., 2011): the sip53-targeting human TP53 (GAC
4.2 Cell culture and drugs UCCAGUGGUAAUCUAC) and siControl-targeting LacZ gene
from Escherichia coli (GUGACCAGCGAAUACCUGU). Stable
HEK-293, HCT116, HEK293-GP, MDAMB231, MDAMB468, genetic manipulation was performed by transduction with
and MCF7 were cultured in DMEM medium (Gibco) supplemented retrovirus-based plasmids as previously described (Girardini et
with 10% fetal bovine serum (FBS, Gibco). MDAMB468 and al., 2011). In brief, retroviral particles were generated by co-
MCF7 were also supplemented with 2 mM glutamine (Gibco). transfection of HEK293-GP cells with the packaging plasmid
H1299 and 4T1 cells were cultured in RPMI 1640 medium pMD2ENV and specific constructs using the calcium phosphate
(Gibco) supplemented with 10% FBS (Gibco). MCF10A cells method. Culture media containing retroviral particles were
were cultured in DMEM:F12 medium (Gibco) supplemented harvested 48 h after transfection, filtered, supplemented with
with 10% FBS (Gibco), 0.5 μg/mL hydrocortisone, 10 μg/mL 10% SFB and 4 μg/mL polybrene (Sigma), and used to transduce
insulin, 20 ng/mL epidermal growth factor, 100 ng/mL choleric target cells. Upon transduction, cells were allowed to recover for
toxin, and 100 U/mL penicillin and streptomycin (Invitrogen). 24 h in fresh media, and cells were selected with puromycin
All media were supplemented with 100 U/mL penicillin and (Sigma).
streptomycin (Invitrogen). Cells were grown in a humidified
incubator at 37°C with 5% CO2. Cell lines were purchased from
the American Type Culture Collection (ATCC), and authenticity 4.5 Gene expression analysis
was documented by standard short tandem Repeat (STR) analysis.
Cells were cultured in a humidified incubator at 37°C with 5% CO2 Total RNA was extracted using TRIzol reagent (Invitrogen) and
and tested periodically for Mycoplasma by 4,6-diamidino-2- subjected to DNase-I (Promega) treatment. RNA was retro-
phenylindole (DAPI) staining and PCR. Doxorubicin (Sigma) transcribed using M-MLV retrotranscriptase (Promega). Real-
was dissolved in DMSO. time PCR was performed using SYBR Green PCR master mix
(Biodynamics) according to the following conditions: 2 min at
95°C for one cycle; 30 s at 95°C, 20 s at 60°C, and 30 s at 72°C for
4.3 Plasmids 40 cycles. The results were analyzed using the comparative ΔCt
method. Values were normalized to GAPDH mRNA levels. The
For p53 transient expression, pCDNA3-p53 was used following real-time PCR primers were used: AKR1B1_F: TGAGTG
(Mantovani et al., 2007). Retroviral plasmids used in this study CCACCCATATCTCA, AKR1B1_R: TGTCACAGACTTGGGGAT
were constructed using the pLPC-GFP vector (Girardini et al., 2011) CA, GAPDH_F: TCTCTGCTCCTCCTGTTC, GAPDH_R: GCC
containing a selectable puromycin cassette. The AKR1B1 coding CAATACGACCAAATCC
sequence was amplified by PCR on cDNA from H1299 cells and
cloned using EcoRI/XhoI sites. The following PCR primers were
used: AKR1B1Eco_F: AAAGAATTCATGGCAAGCCGTCTCCTG 4.6 Western blot
C, AKR1B1Xho_R: AAACTCGAGTCAAAACTCTTCATGGAA
GGG. To generate pAKR1B1luc reporter, a fragment spanning Western blot was performed as previously described (Ibarra
from −1785 to +24 on the AKR1B1 promoter was amplified by et al., 2017). The following primary antibodies were used:

Frontiers in Molecular Biosciences 08 frontiersin.org

Di Benedetto et al. 10.3389/fmolb.2023.1145279

monoclonal anti-p53 (DO-1, Santa Cruz), monoclonal anti- centrifugation, and protein concentration determination,
AKR1B1 (CPTC-AKR1B1-3, DSHB), polyclonal anti-GFP enzymatic activity was quantified in KAB buffer using 50 µg
(Ab290, Abcam), and polyclonal anti-β-actin (A2066, Sigma). proteins from extracts, 175 µM NADPH, and 707 mM D +
HRP-conjugated anti-rabbit (111-035-003, Jackson) and HRP- -xylose as substrate in 500 µL final volume. Initial enzymatic
conjugated anti-mouse (115-035-003, Jackson) secondary rates were followed at 37°C as the decrease in absorbance at
antibodies were used. Chemiluminescence was detected using 340 nm, corresponding to NADPH oxidation. KAB buffer
Amersham ECL Prime Western Blotting Detection Reagent (GE supplemented with NADPH and xylose was used as blank.
Healthcare). For quantitative analysis, the intensity of Western
blot bands was measured using ImageJ (Abràmoff et al., 2004)
and normalized by the intensity of their respective β-actin 4.11 Invasion assay
bands.
For in vitro invasion assays, Transwell chambers (8 µm pore size,
Corning) were pre-coated with 1% Matrigel (Corning) in media
4.7 Luciferase reporter assay without FBS for 1 h at 37°C. 4T1-AKR1B1 or 4T1-GFP cells were
seeded on the upper side in media without FBS, and complete media
H1299 cells were co-transfected with the indicated plasmids and were used in the lower chamber. After incubation for the indicated
pCMV-β-galactosidase (Promega) as a control of transfection time, cells on the upper side of the Transwell were mechanically
efficiency, using Lipofectamine 2000 (Invitrogen). Transfected cells removed, and cells in the lower surface were fixed with 4%
were harvested in Passive Lysis Buffer 1X (Promega). Luciferase p-formaldehyde and stained with Hoechst. Invasive cells were
activity was measured using the Luciferase Assay Reagent (Promega) quantified by fluorescence microscopy using ImageJ FIJI.
in a Multi-Mode Microplate Reader (Synergy 2, BioTek). The values
were normalized relative to β-galactosidase activity.
4.12 In vivo tumor formation

4.8 ChIP assay Procedures involving animals conformed to institutional

guidelines that comply with international laws and policies (the
ChIP was performed as previously described (Borini Etichetti et Council for International Organization of Animal Sciences
al., 2019) on HEK-293 cells, HepG2 cells, H1299 cells transfected (CIOMS) and the International Council for Laboratory Animal
with pCDNA3-p53 or pCDNA3 as a control, or HCT116 cells Science (ICLAS)). All experimental protocols were approved by
treated with 0.5 µM doxorubicin or vehicle (DMSO) for 16 h. the Animal Ethics Committee at the School of Medicine, National
Chromatin was sonicated to 500–800 bp average fragment size University of Rosario (CICUAL-FCM-UNR. Res. 2534/2021). Six-
and pre-cleared for 1 h at 4°C with protein A-Sepharose (GE to eight-week-old female BALB/c mice were obtained from the
Healthcare). Sepharose was removed by centrifugation, and an Center for Comparative Medicine (ICIVET), National University
aliquot of the supernatant was saved as input. Chromatin was of Litoral. Animals were fed with commercial chow and water ad
immunoprecipitated overnight at 4°C with anti-p53 antibody libitum and maintained in a 12-h light and 12-h dark schedule. Mice
(DO1, Santa Cruz) or anti-GFP (sc-365549, Santa Cruz). were randomly divided into two groups (n = 10 per group) and
Immunoprecipitated DNA was analyzed by real-time PCR. subcutaneously injected on the mammary fat pad with 1 × 106 4T1-
Promoter occupancy was calculated as percent of input AKR1B1 or 4T1-GFP cells in PBS. Palpable masses were measured
chromatin using the ΔCt method. The following primers were used: twice a week using calipers, and tumor volume was quantified using
AKR1B1ChIP_F: CGCAACCAATCAGAAGGCTCC, AKR1B1ChIP_R: formula V = (π.D.d2)/6, where D is the largest diameter, and d is the
GGCGCGTACCTTTAAATAGCC smallest diameter of the tumor. Three mice from the 4T1-AKR1B1
group died before the endpoint of the experiment, at days 6, 35, and
37 after injection. The first mouse was excluded from the experiment
4.9 In silico analysis of p53-binding sites analysis, whereas the other two mice were included in the tumor
growth calculation. However, autopsy and metastasis evaluation for
The presence of p53-binding sites on the AKR1B1 promoter was these two mice could not be performed. After completion, mice were
analyzed using JASPAR (Castro-Mondragon et al., 2022). For this sacrificed, and organs were extracted for metastasis evaluation. To
analysis, the region spanning from −1785 to +24 on the facilitate macroscopic visualization of lung metastases, the lungs
AKR1B1 promoter was used. were stained by intra-tracheal injection of India ink followed by
wash as previously described (Miretti et al., 2008).

4.10 Aldose reductase activity

4.13 Proteomic analysis by liquid
Aldose reductase activity was determined using the method chromatography and mass spectrometry
described by Hayman and Kinoshita (1965). In brief, cell extracts
were prepared in KAB buffer (50 mM Hepes pH 7.5; 150 mM NaCl; Whole-cell extracts of 4T1-AKR1B1 and 4T1-GFP cells were
1 mM DTT; and 10 mM MgCl2), supplemented with protease prepared in biological triplicates. Cells were washed with PBS and
inhibitor cocktail (Sigma) and 1mM PMSF. After sonication, resuspended in RIPA buffer. Upon lysis, the supernatants were

Frontiers in Molecular Biosciences 09 frontiersin.org

Di Benedetto et al. 10.3389/fmolb.2023.1145279

recovered by centrifugation for further analysis. Tryptic peptides 4.15 Determination of ROS
were obtained by in-gel digestion (Link and LaBaer, 2009).
Peptide chromatography was performed on an Easy-nLC For ROS determination, a modification of the method reported
1000 nanosystem (Thermo Scientific). Each sample was loaded by Wang and Joseph (1999) was used. Cells were incubated with
into an Acclaim PepMap 100 precolumn (Thermo Scientific) and 5 µM 2′,7′-dichlorofluorescein diacetate (Sigma) in 1% FBS media
eluted into the analytical column (RSLC PepMap C18, 50 cm for 45 min at 37°C in the dark. After washing with PBS, fluorescence
long, 75 µm inner diameter, and 2 µm particle size, Thermo was measured with an excitation wavelength of 488 nm and
Scientific). The mobile phase flow rate was 300 nL/min, using emission wavelength of 525 nm using a Multi-Mode Microplate
water + 0.1% formic acid (solvent A) and acetonitrile + 0.1% Reader (Synergy 2, BioTek). Non-stained cell suspensions were used
formic acid (solvent B). The gradient profile was set as follows: as blank.
5%–35% solvent B for 100 min, 35%–45% solvent B for 20 min,
45%–100% solvent B for 5 min, and 100% solvent B for 15 min.
Mass spectrometry (MS) analysis was performed using a 4.16 Statistical analysis
Q-Exactive mass spectrometer (Thermo Scientific). The full-
scan method employed an m/z 300–1800 mass selection, an All experiments were conducted in triplicates unless otherwise
Orbitrap resolution of 70,000 (at m/z 200), a target automatic indicated. The results were plotted as mean values +/- standard error
gain control (AGC) value of 3e6, and maximum injection times of of the mean using GraphPad Prism7 and statistically analyzed using
100 ms. After the survey scan, the 15 most intense precursor ions a two-tailed t-test or ANOVA followed by a multiple-comparison
were selected for MS/MS fragmentation. Fragmentation was post-test, as appropriate. p-values less than 0.05 were considered
performed with a normalized collision energy of 27 eV, and statistically significant.
MS/MS scans were acquired with a starting mass of m/z 200,
AGC target of 2e5, resolution of 17,500 (at m/z 200), intensity
threshold of 8e3, isolation window of 2.0 m/z units, and Data availability statement
maximum IT of 100 ms. A dynamic exclusion time of 30 s was
used to discriminate against previously selected ions. MS data The mass spectrometry proteomics data presented in this study
were analyzed with Proteome Discoverer 2.4 using standardized are deposited in the ProteomeXchange Consortium (https://www.
workflows. Mass spectra *.raw files were searched against the proteomexchange.org/) via the PRIDE partner repository, with the
database of Mus musculus from UniProt. Precursor and fragment dataset identifier PXD045048.
mass tolerance were set to 10 ppm and 0.02 Da, respectively,
allowing two missed cleavages, carbamidomethylation of
cysteines as a fixed modification, methionine oxidation, and Ethics statement
acetylation N-terminal as a variable modification. Data on
identified and quantified proteins were analyzed with Perseus Ethical approval was not required for the studies on humans
(Tyanova and Cox, 2018). in accordance with the local legislation and institutional
requirements because only commercially available established
cell lines were used. The animal study was approved by the
4.14 Functional enrichment analysis and Animal Ethics Committee, School of Medicine, National
interaction networks University of Rosario. The study was conducted in
accordance with the local legislation and institutional
The DEP list was defined as up- or downregulated proteins requirements.
with an FC of 2 or higher, considering only proteins identified by
two or more unique peptides (p-value ≤ 0.05) resulting from data
processing by Proteome Discoverer. Functional enrichment Author contributions
analysis was performed using the KEGG Mapper tool (https://
www.genome.jp/kegg/mapper/) (Kanehisa and Sato, 2020). A Conceptualization: JG and MM-M; methodology: CD, CB,
heatmap was made to visualize the DEPs in KEGG categories SB, JG, and MM-M; formal analysis and investigation: CD, CB,
using the Matplotlib43 and Seaborn44 packages. All packages NC, AC, EA, GR, JG, and SB; writing—original draft preparation:
were run on Python 3.10. Interaction networks were obtained JG and CD; writing—review and editing: JG, CD, GR, MM-M,
from STRING (Szklarczyk et al., 2021) with a confidence cutoff and SB; funding acquisition: JG; and supervision: JG, MM-M, and
of 0.6 and no extra interactors and loaded onto Cytoscape v3.9.1 GR. All authors contributed to the article and approved the
(Shannon et al., 2003), with a confidence cutoff of 0.6. Network submitted version.
clustering was performed using the MCL algorithm set to a
granularity parameter of 2.5. For network analysis, the DEP list
was extended considering proteins with log 2FC ≥ 0.85 or Funding
log2 FC ≤ −0.85 (p-value ≤ 0.05), in order to increase the
number of interactors. Proteins without interaction partners This work was supported by grants from Instituto Nacional del
within the network (singletons) were omitted from the Cáncer (Asistencia Financiera III, 2015) to JG and from
visualization. ANPCyT—MINCyT (PICT-1221) to JG.

Frontiers in Molecular Biosciences 10 frontiersin.org

Di Benedetto et al. 10.3389/fmolb.2023.1145279

Acknowledgments Publisher’s note

The authors thank Cecilia Farre and Gustavo Chapo from All claims expressed in this article are solely those of the authors and
CIPReB (Centro de Investigación y Producción de Reactivos do not necessarily represent those of their affiliated organizations, or
biológicos, UNR) for helping with mouse housing and those of the publisher, the editors, and the reviewers. Any product that
monitoring. They also thank Mauro Ibañez for helping with may be evaluated in this article, or claim that may be made by its
the visualization of functional enrichment data. manufacturer, is not guaranteed or endorsed by the publisher.

Conflict of interest Supplementary material

The authors declare that the research was conducted in the The Supplementary Material for this article can be found online
absence of any commercial or financial relationships that could be at: https://www.frontiersin.org/articles/10.3389/fmolb.2023.1145279/
construed as a potential conflict of interest. full#supplementary-material

References
Abràmoff, M. D., Magalhães, P. J., and Ram, S. J. (2004). Image processing with Ibarra, M., Borini Etichetti, C., DiBenedetto, C., Rosano, G., Margarit, E., Del Sal, G.,
ImageJ. Biophot. Int. 11, 36–42. et al. (2017). Dynamic regulation of Pin1 expression and function during zebrafish
development. PLoS One 12, e0175939. doi:10.1371/journal.pone.0175939
Aslakson, C. J., and Miller, F. R. (1992). Selective events in the metastatic process
defined by analysis of the sequential dissemination of subpopulations of a mouse Kanehisa, M., and Sato, Y. (2020). KEGG Mapper for inferring cellular functions from
mammary tumor. Cancer Res. 52, 1399–1405. doi:10.1002/0471142735.im2002s39 protein sequences. Protein Sci. 29, 28–35. doi:10.1002/pro.3711
Bensaad, K., Tsuruta, A., Selak, M. A., Vidal, M. N., Nakano, K., Bartrons, R., et al. Kastenhuber, E. R., and Lowe, S. W. (2017). Putting p53 in context. Cell 170,
(2006). TIGAR, a p53-inducible regulator of glycolysis and apoptosis. Cell 126, 107–120. 1062–1078. doi:10.1016/j.cell.2017.08.028
doi:10.1016/j.cell.2006.05.036
Kim, J.-Y., Kim, H.-J., Jung, C.-W., Choi, B.-I., Lee, D.-H., and Park, M.-J. (2021).
Borini Etichetti, C., Di Benedetto, C., Rossi, C., Baglioni, M., Bicciato, S., Del Sal, G., PARK7 maintains the stemness of glioblastoma stem cells by stabilizing epidermal
et al. (2019). Isoprenylcysteine carboxy methyltransferase (ICMT) is associated with growth factor receptor variant III. Oncogene 40, 508–521. doi:10.1038/s41388-020-
tumor aggressiveness and its expression is controlled by the p53 tumor suppressor. 01543-1
J. Biol. Chem. 294, 5060–5073. doi:10.1074/jbc.RA118.006037
Kinoshita, J. H. (1990). A thirty year journey in the polyol pathway. Exp. Eye Res. 50,
Borini Etichetti, C. M., Arel Zalazar, E., Cocordano, N., and Girardini, J. (2020). 567–573. doi:10.1016/0014-4835(90)90096-D
Beyond the mevalonate pathway: control of post-prenylation processing by mutant p53.
Koboldt, D. C., Fulton, R. S., McLellan, M. D., Schmidt, H., Kalicki-Veizer, J.,
Front. Oncol. 10, 595034. doi:10.3389/fonc.2020.595034
McMichael, J. F., et al. (2012). Comprehensive molecular portraits of human breast
Bresson, E., Lacroix-Pepin, N., Boucher-Kovalik, S., Chapdelaine, P., and Fortier, M. tumours. Nature 490, 61–70. doi:10.1038/nature11412
A. (2012). The prostaglandin F synthase activity of the human aldose reductase
Lelekakis, M., Moseley, J. M., Martin, T. J., Hards, D., Williams, E., Ho, P., et al.
AKR1B1 brings new lenses to look at pathologic conditions. Front. Farmacol. 3, 98.
(1999). A novel orthotopic model of breast cancer metastasis to bone. Clin.
doi:10.3389/fphar.2012.00098
Exp. Metastasis 17, 163–170. doi:10.1023/A:1006689719505
Castro-Mondragon, J. A., Riudavets-Puig, R., Rauluseviciute, I., Lemma, R. B., Turchi,
Link, A. J., and LaBaer, J. (2009). In-gel trypsin digest of gel-fractionated proteins.
L., Blanc-Mathieu, R., et al. (2022). Jaspar 2022: the 9th release of the open-access
Cold Spring Harb. Protoc. 2009, pdb.prot5110. doi:10.1101/pdb.prot5110
database of transcription factor binding profiles. Nucleic Acids Res. 50, D165–D173.
doi:10.1093/nar/gkab1113 Mantovani, F., Tocco, F., Girardini, J., Smith, P., Gasco, M., Lu, X., et al. (2007). The
prolyl isomerase Pin1 orchestrates p53 acetylation and dissociation from the apoptosis
Cerami, E., Gao, J., Dogrusoz, U., Gross, B. E., Sumer, S. O., Aksoy, B. A., et al. (2012).
inhibitor iASPP. Nat. Struct. Mol. Biol. 14, 912–920. doi:10.1038/nsmb1306
The cBio cancer genomics portal: an open platform for exploring multidimensional
cancer genomics data. Cancer Discov. 2, 401–404. doi:10.1158/2159-8290.CD-12-0095 Miretti, S., Roato, I., Taulli, R., Ponzetto, C., Cilli, M., Olivero, M., et al. (2008). A
mouse model of pulmonary metastasis from spontaneous osteosarcoma monitored in
Cheung, E. C., and Vousden, K. H. (2022). The role of ROS in tumour
vivo by luciferase imaging. PLoS One 3, e1828. doi:10.1371/journal.pone.0001828
development and progression. Nat. Rev. Cancer 22, 280–297. doi:10.1038/
s41568-021-00435-0 Moon, S.-H., Huang, C.-H., Houlihan, S. L., Regunath, K., Freed-Pastor, W. A.,
Morris, J. P., et al. (2019). p53 represses the mevalonate pathway to mediate tumor
Curtis, C., Shah, S. P., Chin, S.-F., Turashvili, G., Rueda, O. M., Dunning, M. J., et al.
suppression. Cell 176, 564–580. doi:10.1016/j.cell.2018.11.011
(2012). The genomic and transcriptomic architecture of 2,000 breast tumours reveals
novel subgroups. Nature 486, 346–352. doi:10.1038/nature10983 Napoli, M., and Flores, E. R. (2017). The p53 family orchestrates the regulation of
metabolism: physiological regulation and implications for cancer therapy. Br. J. Cancer.
Ghandi, M., Huang, F. W., Jané-Valbuena, J., Kryukov, G. V., Lo, C. C., McDonald, E.
116, 149–155. doi:10.1038/bjc.2016.384
R., 3rd, et al. (2019). Next-generation characterization of the cancer cell line
Encyclopedia. Nature 569, 503–508. doi:10.1038/s41586-019-1186-3 Nusinow, D. P., Szpyt, J., Ghandi, M., Rose, C. M., McDonald, E. R., 3rd, Kalocsay, M.,
et al. (2020). Quantitative proteomics of the cancer cell line Encyclopedia. Cell 180,
Girardini, J. E., Napoli, M., Piazza, S., Rustighi, A., Marotta, C., Radaelli, E., et al.
387–402. doi:10.1016/j.cell.2019.12.023
(2011). A pin1/mutant p53 Axis promotes aggressiveness in breast cancer. Cancer Cell
20, 79–91. doi:10.1016/j.ccr.2011.06.004 Pastel, E., Pointud, J. C., Volat, F., Martinez, A., and Lefrançois-Martinez, A. M.
(2012). Aldo-keto reductases 1B in endocrinology and metabolism. Front. Pharmacol. 3,
Győrffy, B. (2021). Survival analysis across the entire transcriptome identifies
148. doi:10.3389/fphar.2012.00148
biomarkers with the highest prognostic power in breast cancer. Comput. Struct.
Biotechnol. J. 19, 4101–4109. doi:10.1016/j.csbj.2021.07.014 Pavlova, N. N., and Thompson, C. B. (2016). The emerging hallmarks of cancer
metabolism. Cell Metab. 23, 27–47. doi:10.1016/j.cmet.2015.12.006
Hainaut, P., and Pfeifer, G. P. (2016). Somatic TP53 mutations in the era of genome
sequencing. Cold Spring Harb. Perspect. Med. 6, a026179. doi:10.1101/cshperspect. Röhrig, F., and Schulze, A. (2016). The multifaceted roles of fatty acid synthesis in
a026179 cancer. Nat. Rev. Cancer 16, 732–749. doi:10.1038/nrc.2016.89
Hanahan, D., and Weinberg, R. A. (2011). Hallmarks of cancer: the next generation. Sánchez-Gómez, F. J., Díez-Dacal, B., García-Martín, E., Agúndez, J. A. G.,
Cell 144, 646–674. doi:10.1016/j.cell.2011.02.013 Pajares, M. A. M. A., Pérez-Sala, D., et al. (2016). Detoxifying enzymes at the cross-
roads of inflammation, oxidative stress, and drug hypersensitivity: role of
Hayman, S., and Kinoshita, J. H. (1965). Isolation and properties of lens aldose
glutathione transferase P1-1 and aldose reductase. Front. Pharmacol. 7,
reductase. J. Biol. Chem. 240, 877–882. doi:10.1016/S0021-9258(17)45256-2
237–239. doi:10.3389/fphar.2016.00237
Hu, T., Chen, X., Lu, S., Zeng, H., Guo, L., and Han, Y. (2022). Biological role and
Schwab, A., Siddiqui, A., Vazakidou, M. E., Napoli, F., Bottcher, M., Menchicchi, B.,
mechanism of lipid metabolism reprogramming related gene ECHS1 in cancer. Technol.
et al. (2018). Polyol pathway links glucose metabolism to the aggressiveness of cancer
Cancer Res. Treat. 21, 15330338221140655. doi:10.1177/15330338221140655
cells. Cancer Res. 78, 1604–1618. doi:10.1158/0008-5472.CAN-17-2834

Frontiers in Molecular Biosciences 11 frontiersin.org

Di Benedetto et al. 10.3389/fmolb.2023.1145279

Schwartzenberg-Bar-Yoseph, F., Armoni, M., and Karnieli, E. (2004). The tumor Tyanova, S., and Cox, J. (2018). “Perseus: A bioinformatics platform for integrative
suppressor p53 down-regulates glucose transporters GLUT1 and GLUT4 gene analysis of proteomics data in cancer research,” in Methods in molecular biology
expression. Cancer Res. 64, 2627–2633. doi:10.1158/0008-5472.CAN-03-0846 (Humana Press Inc.), 133–148. doi:10.1007/978-1-4939-7493-1_7
Shannon, P., Markiel, A., Ozier, O., Baliga, N. S., Wang, J. T., Ramage, D., et al. (2003). Venkatesan, R., Sah-Teli, S. K., Awoniyi, L. O., Jiang, G., Prus, P., Kastaniotis, A. J.,
Cytoscape: A software environment for integrated models of biomolecular interaction et al. (2014). Insights into mitochondrial fatty acid synthesis from the structure of
networks. Genome Res. 13, 2498–2504. doi:10.1101/gr.1239303 heterotetrameric 3-ketoacyl-ACP reductase/3R-hydroxyacyl-CoA dehydrogenase. Nat.
Commun. 5, 4805. doi:10.1038/ncomms5805
Shen, Y., Zhong, L., Johnson, S., and Cao, D. (2011). Human aldo-keto reductases
1B1 and 1B10: A comparative study on their enzyme activity toward electrophilic Wang, B., Abraham, N., Gao, G., and Yang, Q. (2016). Dysregulation of autophagy
carbonyl compounds. Chem. Biol. Interact. 191, 192–198. doi:10.1016/j.cbi.2011.02.004 and mitochondrial function in Parkinson’s disease. Transl. Neurodegener. 5, 19. doi:10.
1186/s40035-016-0065-1
Silwal-Pandit, L., Vollan, H. K. M., Chin, S. F., Rueda, O. M., McKinney, S., Osako, T.,
et al. (2014). TP53 mutation spectrum in breast cancer is subtype specific and has Wang, H., and Joseph, J. A. (1999). Quantifying cellular oxidative stress by
distinct prognostic relevance. Clin. Cancer Res. 20, 3569–3580. doi:10.1158/1078-0432. dichlorofluorescein assay using microplate reader. Free Radic. Biol. Med. 27,
CCR-13-2943 612–616. doi:10.1016/s0891-5849(99)00107-0
Srivastava, S. K., Ramana, K. V., and Bhatnagar, A. (2005). Role of aldose reductase Wu, J. W., Preuss, C., Wang, S. P., Yang, H., Ji, B., Carter, G. W., et al. (2017a).
and oxidative damage in diabetes and the consequent potential for therapeutic options. Epistatic interaction between the lipase-encoding genes Pnpla2 and Lipe causes
Endocr. Rev. 26, 380–392. doi:10.1210/er.2004-0028 liposarcoma in mice. PLOS Genet. 13, e1006716. doi:10.1371/journal.pgen.1006716
Szklarczyk, D., Gable, A. L., Nastou, K. C., Lyon, D., Kirsch, R., Pyysalo, S., et al. Wu, X., Li, X., Fu, Q., Cao, Q., Chen, X., Wang, M., et al. (2017b). AKR1B1 promotes
(2021). The STRING database in 2021: customizable protein–protein networks, and basal-like breast cancer progression by a positive feedback loop that activates the EMT
functional characterization of user-uploaded gene/measurement sets. Nucleic Acids Res. program. J. Exp. Med. 214, 1065–1079. doi:10.1084/jem.20160903
49, D605–D612. doi:10.1093/nar/gkaa1074
Yerlikaya, A., and Erin, N. (2008). Differential sensitivity of breast cancer and
Tammali, R., Srivastava, S. K., and Ramana, K. V. (2012). Targeting aldose reductase melanoma cells to proteasome inhibitor Velcade. Int. J. Mol. Med. 22, 817–823.
for the treatment of cancer. Curr. Cancer Drug Targets 11, 560–571. doi:10.2174/ doi:10.3892/ijmm_00000090
156800911795655958
Zhang, C., Lin, M., Wu, R., Wang, X., Yang, B., Levine, A. J., et al. (2011). Parkin, a
Tang, W. H., Martin, K. A., and Hwa, J. (2012). Aldose reductase, oxidative stress, and p53 target gene, mediates the role of p53 in glucose metabolism and the Warburg effect.
diabetic mellitus. Front. Pharmacol. 3, 87. doi:10.3389/fphar.2012.00087 Proc. Natl. Acad. Sci. U. S. A. 108, 16259–16264. doi:10.1073/pnas.1113884108

Frontiers in Molecular Biosciences 12 frontiersin.org