Abstract
In humans, a neomorphic isocitrate dehydrogenase mutation (idh-1neo) causes increased levels of cellular D-2-hydroxyglutarate (D-2HG), a proposed oncometabolite. However, the physiological effects of increased D-2HG and whether additional metabolic changes occur in the presence of an idh-1neo mutation are not well understood. We created a Caenorhabditis elegans model to study the effects of the idh-1neo mutation in a whole animal. Comparing the phenotypes exhibited by the idh-1neo to ∆dhgd-1 (D-2HG dehydrogenase) mutant animals, which also accumulate D-2HG, we identified a specific vitamin B12 diet-dependent vulnerability in idh-1neo mutant animals that leads to increased embryonic lethality. Through a genetic screen, we found that impairment of the glycine cleavage system, which generates one-carbon donor units, exacerbates this phenotype. In addition, supplementation with alternate sources of one-carbon donors suppresses the lethal phenotype. Our results indicate that the idh-1neo mutation imposes a heightened dependency on the one-carbon pool and provides a further understanding of how this oncogenic mutation rewires cellular metabolism.
Introduction
Increased levels of D-2-hydroxyglutarate (D-2HG), a metabolite derived from the structurally similar hub metabolite alpha-ketoglutarate (αKG), are associated with multiple disorders, indicating that tight regulation of D-2HG is important (1, 2). For instance, D-2-hydroxyglutaric aciduria, a rare inborn error of metabolism, is associated with elevated D-2HG levels due to loss-of-function mutations in the D-2HG dehydrogenase enzyme (3). This inborn error of metabolism often results in neurological dysfunctions and delayed development. Previously, we found that loss of the Caenorhabditis elegans D-2HG dehydrogenase (dhgd-1) causes a high rate of embryonic lethality due to reduced ketone body production (4). Additionally, we found that dhgd-1 activity is necessary for the regulation of the propionate shunt, an alternate vitamin B12-independent breakdown pathway for this short chain fatty acid (5). In this shunt, the enzymes DHGD-1 and HPHD-1 are coupled via D-2HG metabolism: HPHD-1 transfers a hydride from 3HP to α-ketoglutarate (αKG), producing D-2HG, whereas DHGD-1 oxidizes D-2HG back to αKG (4, 5) (Fig 1A). The propionate shunt is transcriptionally repressed in the presence of vitamin B12 (6). Vitamin B12 rescues the embryonic lethality of ∆dhgd-1 mutants by generating energy via the canonical propionate degradation pathway, alleviating the need for ketone bodies to distribute an energy source across tissues (4) (Fig 1A).
D-2HG is also known as an oncometabolite and is linked to various cancers. D-2HG accumulates due to neomorphic mutations in either one of two isocitrate dehydrogenase (IDH) enzymes (IDH1 and IDH2). These mutations primarily affect catalytic arginine residues (7, 8, 9) and are associated with the development of cancers such as glioma, cholangiocarcinoma, and AML (10, 11, 12).
Neomorphic mutations in IDH1 and IDH2 enzymes lead to abnormal D-2HG production from αKG (13), thereby disrupting cell function. Effects of D-2HG are multifaceted and can drive cancer development by several different mechanisms (1, 14). D-2HG acts as a potent competitive inhibitor of αKG-dependent enzymes, including histone demethylases and hypoxia-inducible factor prolyl hydroxylase, often leading to dysregulated oncogene expression (15). Abnormal D-2HG production also disturbs the balance between NADPH and NADP+, crucial for cellular redox equilibrium (16). This disruption can cause oxidative stress, leading to DNA damage. High levels of D-2HG have also been shown to inhibit succinate dehydrogenase (17) and αKG-dependent transaminases (16), disrupt chromosomal topology (18), and activate the mTOR pathway (19). D-2HG also affects the immune system, particularly T cells, potentially creating a tumor-friendly environment by suppressing the immune response. Malignant cells with IDH mutations release D-2HG, which can suppress T-cell function by inhibiting lactate dehydrogenase and disrupt other metabolic pathways (20, 21, 22). Increased levels of D-2HG caused by the inhibition of D-2-hydroxyglutarate dehydrogenase activity have also been associated with different cancers (23, 24, 25). Whereas many effects of D-2HG are well-documented, the complete implications of dysregulated D-2HG metabolism remain unclear. Its versatile effects range from supporting oncometabolism to causing developmental and psychomotor defects in D-2-hydroxyglutaric aciduria patients. Understanding the diverse toxic effects of D-2HG is crucial for unraveling disease progression mechanisms and developing new treatments.
To gain a better understanding of how D-2HG impacts cellular metabolic function, we generated C. elegans idh-1neo mutant animals to use as a comparative model for studying the effects of increased D-2HG levels. We find that whereas some shifts in metabolism are shared with what we found previously in our studies of ∆dhgd-1 mutant animals (4), differences exist. These differences led us to uncover a unique diet-dependent, vitamin B12-induced vulnerability in idh-1neo mutant animals. Whereas vitamin B12 rescues embryonic lethality in ∆dhgd-1 mutant animals, it exacerbates lethality of idh-1neo mutant animals. We find that this difference is due to decreased one carbon metabolism in idh-1neo mutant animals. Overall, our results provide a further understanding of how the idh-1neo oncogenic mutation may rewire cellular metabolism.
Results
C. elegans with neomorphic idh-1 mutation accumulate D-2HG
We previously found that when the function of the C. elegans D-2HG dehydrogenase, dhgd-1, is disrupted, there is a marked increase in D-2HG levels in the animals (Fig 1B, Model 1) (4). Seeking to further understand the metabolic implications of D-2HG accumulation, we aimed to increase D-2HG levels through a distinct mechanism; by introducing an exogenous D-2HG-producing enzyme (Fig 1B, Model 2). Neomorphic mutations in IDH, whether cytosolic (IDH1) or mitochondrial (IDH2), alter enzyme function; rather than converting isocitrate to αKG, these mutant enzymes convert αKG to D-2HG (Fig 1C). We separately introduced four idh alleles with missense mutations that mirror human neomorphic variants into C. elegans (Fig 1D and E), whereas keeping the endogenous WT idh genes intact. Using these mutant strains, we assessed 2HG (D- and L-2HG) accumulation in the animals by gas chromatography—mass spectrometry (GC-MS). Whereas neither idh-2 allele resulted in 2HG accumulation, animals harboring an idh-1 alleles did show an increased accumulation of 2HG, with the highest levels found in animals expressing the idh-1(R156C) allele (Fig 1F). Approximately 50% of idh-1(R156C) animals also showed developmental abnormalities, including dilations in the excretory system, larval lethality, and a smaller proportion of embryonic lethality, suggesting systemic disruptions to animal physiology (Fig S1A and B and Table S1). Therefore, we chose this strain for detailed exploration, and hereafter refer to it as idh-1neo.
2HG exists as either the D-2HG or L-2HG enantiomer, and neomorphic IDH mutations specifically cause production of D-2HG (27). Using a specific derivatization technique to distinguish 2HG enantiomers (28), we confirmed that idh-1neo animals accumulate D-2HG (Fig 1G). These combined results show that we have generated a new model for D-2HG accumulation in C. elegans, distinct from that in ∆dhgd-1 animals (4). Using these two models, we went on to further understand the metabolic implications of D-2HG accumulation.
Vitamin B12 supplementation increases D-2HG levels and causes embryonic lethality in idh-1neo mutants
We next investigated whether metabolites other than D-2HG change in abundance in idh-1neo animals. GC-MS metabolomics revealed that, much like ∆dhgd-1 animals, idh-1neo animals exhibited elevated levels of 3-hydroxypropionate (3HP) and β-alanine, along with reduced levels of αKG and aspartate (Fig 2A) (4).
In ∆dhgd-1 mutants, 3HP accumulates because D-2HG inhibits HPHD-1, an enzyme that produces D-2HG whereas oxidizing 3HP in the propionate shunt pathway (4, 29) (Fig 2B). To determine if 3HP accumulation in idh-1neo also originates from this shunt pathway, we supplemented the animals with vitamin B12. Vitamin B12 transcriptionally inhibits the propionate shunt pathway whereas promoting the activity of the canonical, B12-dependent propionate degradation pathway (5, 6) (Fig 1A). We reasoned that if suppressed HPHD-1 activity is the cause of 3HP accumulation, then inhibiting the entire shunt pathway should prevent it. Indeed, vitamin B12 supplementation led to reduced 3HP levels in idh-1neo animals, which also occurred in ∆dhgd-1 mutant animals (Fig 2C and D). Interestingly, in contrast to ∆dhgd-1 mutant animals, vitamin B12 supplementation significantly increased the rate of embryonic but not larval lethality in the F1 generation of idh-1neo animals (Fig 2E and Table S1) (4). We hypothesized that because of vitamin B12 suppresses the expression of shunt pathway genes, including dhgd-1, its supplementation may hinder DHGD-1 dependent D-2HG recycling, thereby further elevating D-2HG levels in idh-1neo animals (Fig 2F). Indeed, adding vitamin B12 to the diet of the idh-1neo significantly increased their D-2HG levels (Figs 2G and S2). To test this hypothesis further, we asked if suppressing dhgd-1 expression would elevate D-2HG in idh-1neo animals. As predicted, dhgd-1 RNAi was sufficient to drive further increase in D-2HG levels in idh-1neo animals (Fig 2H). Importantly, dhgd-1 RNAi also led to 100% penetrant embryonic lethality among the F1 generation of idh-1neo animals (Fig 2I). In contrast, hphd-1 RNAi did not cause embryonic lethality, further demonstrating that lack of 3HP degradation is not linked to this phenotype (Fig S3) (4).
The opposite response to vitamin B12 supplementation highlighted key differences between the two models of D-2HG accumulation. The embryonic lethality observed in ∆dhgd-1 animals arises from a lack of energy source (ketone bodies) and can be rescued by vitamin B12, which activates an alternative energy production pathway (4). In contrast, embryonic lethality of idh-1neo animals is induced by vitamin B12 and cannot be mitigated by ketone body supplementation (Fig S4). We therefore conclude that idh-1neo mutation causes embryonic lethality through a different molecular mechanism.
Knockdown of the glycine cleavage system (GCS) exacerbates lethality of idh-1neo animals supplemented with vitamin B12
To identify the molecular mechanism underlying the lethality of idh-1neo animals in the presence of vitamin B12, we conducted a reverse genetic screen. We used an RNAi library targeting 2,104 predicted metabolic genes (30) to identify those that are essential for idh-1neo animals but not required for WT C. elegans survival in the presence of vitamin B12 (Fig 3A). The screen identified five metabolic genes whose depletion is specifically lethal to idh-1neo animals (Figs 3B and S5). Among these, two genes, T04A8.7 and W07E11.1, encode a glycogen branching enzyme and glutamate synthase, respectively. The other three identified genes—gldc-1, gcst-1, and gcsh-1—all belonging to the GCS (31). Two other GCS genes, gcsh-2 and dld-1 were not identified as “hits.” gcsh-2 is associated with the same reaction as gcsh-1, indicating that the latter encodes an active enzyme (31). dld-1 functions in other metabolic processes, particularly in lactate/pyruvate metabolism, and confers embryonic lethality when knocked down in WT animals (32). Given the strong enrichment for GCS in our screen results, we next considered possible connections between GCS, vitamin B12, and idh-1neo.
idh-1neo mutation confers sensitivity to perturbations of one-carbon metabolism
The GCS breaks down glycine, thereby generating ammonia, carbon dioxide, and reducing NAD+ to NADH, whereas also methylating tetrahydrofolate, a one-carbon (1C) unit donor used for different biosynthetic reactions (Fig 4A). 1C metabolism, similar to the canonical propionate breakdown pathway, also depends on vitamin B12: in the methionine/S-adenosylmethionine (Met/SAM) cycle, METR-1 (methionine synthase) methylates homocysteine to regenerate methionine using vitamin B12 as a cofactor. The Met/SAM cycle utilises 1C units provided by the enzyme methylenetetrahydrofolate reductase MTHF-1 (33) (Fig 4A). Both the GCS and the Met/SAM cycle influence the 1C pool of methylene tetrahydrofolate: GCS contributes to its synthesis, whereas the Met/SAM cycle used it. Therefore, we hypothesized that idh-1neo animals are sensitive to depletion of the 1C pool (Fig 4B). To test this hypothesis, we supplemented B12-treated idh-1neo animals with formate, an alternative 1C donor (34). The highest doses of supplemented formate somewhat slowed the development of P0 animals but restored the survival of idh-1neo embryos to WT levels on a regular diet of E. coli OP50 as well as the diet of RNAi-competent E. coli HT115 (Figs 4C and S6A and B). The alternative 1C donor, serine, also rescued embryonic lethality of idh-1neo animals, but only when fed an E. coli OP50 diet (Figs 4D and S7A and B). Furthermore, we posited that if vitamin B12 induces lethality in idh-1neo animals by depleting the 1C pool via its utilization in the Met/SAM cycle, then suppressing Met/SAM cycle genes in idh-1neo should prevent this depletion and restore availability of 1C units for other reactions (Fig 4A). Indeed, RNAi depletion of mthf-1 and sams-1 (SAM synthetase) rescued the embryonic lethality of idh-1neo animals supplemented with vitamin B12 (Figs 4E and S8). These findings demonstrate that lack of 1C units contributes to the embryonic lethality observed in idh-1neo animals.
Discussion
By comparing two models of D-2HG accumulation in C. elegans, we have gained deeper insight into the metabolic perturbations caused by D-2HG in a whole animal. Similarities between ∆dhgd-1 and idh-1neo include the perturbed function of the propionate shunt enzyme HPHD-1, evident from an increase in levels of its substrate 3HP. Other similarities include elevated β-alanine and reduced αKG. The differences in the metabolic phenotypes of the two models include changes in lysine, 2-aminoadipate, and glutarate levels, and can be linked to the compartmentalization of D-2HG production and the different subcellular origins of D-2HG: DHGD-1 recycles D-2HG produced by HPHD-1 in mitochondria, whereas IDH-1neo generates D-2HG in the cytosol. DHGD-1 dysfunction is thus more likely to affect mitochondrial enzymes whereas IDH-1neo may have a stronger impact on cytosolic metabolism. Consistent with this theory, 3HP levels, indicative of HPHD-1 inhibition, are several-fold higher in ∆dhgd-1 mutants than in idh-1neo animals. In further support of subcellular stratification, mitochondrial lysine degradation pathway intermediates (lysine and 2-aminoadipate) change levels in ∆dhgd-1 mutants, but not in idh-1neo animals (Fig 2A). These lysine levels, however, become perturbed in idh-1neo when vitamin B12, a transcriptional suppressor of dhgd-1, is supplemented (Fig 2C).
1C units in the form of methylated tetrahydrofolate are essential metabolic intermediates used for nucleotide biosynthesis and various methylation reactions (35). A lack of these building blocks results in embryonic lethality (34). Formate, a one-carbon donor exchanged between mitochondria and cytosol, has been demonstrated to rescue these detrimental effects (36). Our results show that idh-1neo C. elegans rely on GCS to supply one-carbon units. We propose that the metabolic rewiring caused by the idh-1neo mutation reduces the availability of methylated tetrahydrofolate. This limitation, in turn, causes sensitivity of idh-1neo to vitamin B12 and GCS knockdown, both of which can drain the 1C pool (Fig 4F). We propose that a lack of 1C units in idh-1neo can impede pyrimidine biosynthesis via thymidylate synthase tyms-1, which uses 1C units to generate dTMP. Supporting this hypothesis, RNAi of tyms-1 causes embryonic lethality (37, 38, 39). WT C. elegans can generate 1C via cytosolic serine hydroxymethyltransferase MEL-32, whose loss causes embryonic lethality (31, 33, 40). Why would the MEL-32 route for 1C unit generation not be available in idh-1neo animals? One possibility is inhibition of this pathway through accumulated D-2HG. The phosphoglycerate dehydrogenase C31C9.2 functions upstream of MEL-32 (31), and its human ortholog PHGDH was found to produce D-2HG (41). A recent study demonstrated that D-2HG accumulation in ∆dhgd-1 animals suppresses the activity of the D-2HG-producing enzyme HPHD-1 (29). A similar mechanism of end-product inhibition could cause the excess D-2HG produced by idh-1neo to suppress C31C9.2 activity, limiting the downstream generation of 1C by MEL-32.
Overall, our results uncover metabolic perturbations induced by the idh-1neo mutation and highlight the differences in the pathogenicity mechanism of idh-1neo and ∆dhgd-1 models. Whereas both mutants accumulate D-2HG and incur embryonic lethality, the ∆dhgd-1 phenotype is caused by a lack of ketone bodies, while idh-1neo suffers from a 1C deficiency. Comparing the two models offers a unique tool for mechanistic insight. These findings may help navigate metabolic reprogramming that occurs in IDH-driven oncogenic transformations. Whereas our results have focused on how the neomorphic idh-1 mutation affects the developing embryo, proliferating cancer cells also have been shown to have increased demand for 1C units, for instance, to synthesize nucleosides (34, 35). Thus, we can speculate that cancers with mutated IDH1 may be increasingly sensitive to depletion of the 1C pool, also. Future studies may explore 1C metabolism as a potential target in the therapy of cancers with the IDH1mutation.
Materials and Methods
Bacterial strains
E. coli HT115, E. coli OP50 (xu363) (42), and E. coli OP50 from Caenorhabditis Genetics Center (CGC) were cultured overnight in Luria-Bertani Broth (Miller) at 37°C, plated, and incubated overnight on assay plates before adding C. elegans larvae. For RNAi experiments, E. coli HT115 was used, and assay plates were supplemented with μg/ml 50 ampicillin and 1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG).
C. elegans cultures
C. elegans strains (Table 1) were maintained at 20°C on nematode growth medium (NGM) seeded with E. coli HT115 or OP50. All experiments were performed using an E. coli OP50 diet, unless specified otherwise. Supplements were added to NGM agar as specified. Vitamin B12 (adenosylcobalamin) was used at 64 nM throughout. N2 strain was obtained from CGC and mutant strains were constructed as described below.
Constructing C. elegans strains
Transgenic C. elegans strains with neomorphic mutations in idh-1 and idh-2 were created by inserting a mutated gene in an intergenic region on chromosome II at position 8420158..8420158 using Mos1-mediated single copy insertion (MosSCI) technique (44). We used expression of an added allele to ensure that endogenous idh-1 remains functional because WT IDH1 activity was demonstrated to be necessary for efficient D-2HG production in cells with monoallelic neomorph mutations of IDH1 (45). WT idh genes, together with their promotor regions, were amplified from C. elegans genomic DNA using a high-fidelity polymerase. Neomorphic missense mutations were introduced using QuikChange Lightning site-directed mutagenesis kit (Agilent). C. elegans strain EG6699 with mos1 site on chromosome II was used for a direct insertion. 50 animals in the L4/young adult stages were injected with a mix of vectors carrying transgene, Mos1 transposase and selection markers. Injection mix contained 2.5 μg/ml of pCFJ90 (Pmyo-2::mCherry), 5 μg/ml pCFJ104 (Pmyo-3::mCherry), 10 μg/ml pGH8 (Prab-3::mCherry), 50 μg/ml of pCFJ601 (Peft-3::Mos1 transposase), 10 μg/ml of pMA122 (Phsp16.41::peel-1), and pCFJ150 with mutated idh sequence. Progeny of individual P0 animals were allowed to starve at 25°C and heat shocked at 34°C for 2 h in a water bath. After 4 h of recovery at 20°C WT moving animals without mCherry expression were picked onto individual plates. Resulting lines with full transmittance were verified for transgene integration by PCR.
C. elegans synchronization
Synchronized L1 populations were obtained by treating gravid adult animals with 1% sodium hypochlorite solution buffered with sodium hydroxide. Released embryos were washed with M9 buffer four times and incubated on a rocker for 18–20 h.
GC-MS metabolomics
Targeted quantification of metabolites by GC-MS was performed as described previously (4). Gravid adult animals were washed three times with filter-sterilized saline (0.9% NaCl). 50 μl of washed animal pellet were transferred into a FastPrep tube (MP Biomedicals), flash frozen in ethanol/dry ice bath and stored at −80°C. Samples were homogenized in 1 ml of 80% cold methanol with 0.5 ml of acid-washed glass beads (Sigma-Aldrich) using FastPrep24 bead beater (MP-Bio). Supernatant was cleared by centrifugation for 10 min at 10,000g. For each sample, 250 μl of cleared extract were transferred into a glass insert and dried under vacuum. Dry residues were derivatized with 20 μl of 20 mg/ml methoxyamine hydrochloride (Sigma-Aldrich) in pyridine for 1 h at 37°C. This step was followed by adding 50 μl of N-methyl-N-(trimethylsilyl) trifluoroacetamide (Sigma-Aldrich) and a subsequent 3-h incubation at 37°C. After additional 5-h RT incubation, the samples were analyzed on an Agilent single quadrupole mass spectrometer 5977B coupled with gas chromatograph 7890B. HP-5MS Ultra Inert capillary column (30 m × 0.25 mm × 0.25 μm) was used with a constant 1 ml/min flow rate of helium gas. Temperature settings were as follows: inlet at 230°C, transfer line at 280°C, MS source at 230°C, and quadrupole at 150°C. A 1 μl sample was injected in split mode with a 5 ml/min split flow. The initial oven temperature was 80°C, rising to 310°C at a 5°C/min rate. MS parameters included 3 scans/s across a 30–500 m/z range and an electron impact ionization energy of 70 eV. Each metabolite’s identification relied on its retention time, a quantifier ion, and two qualifier ions, all manually selected using a reference compound. Peak integration and peak area quantification were executed using Agilent’s MassHunter software (v10.1). Blank subtraction and normalization relative to total quantified metabolites were performed using R software.
Relative quantification of D- and L-2HG
A previously published method (28) was adapted to differentiate the D- and L-enantiomers of 2HG. Initially, 300 μl of C. elegans metabolite extract were dried in glass inserts. 50 μl of R-(-)-butanol and 5 μl of 12N hydrochloric acid were then introduced into each insert and heated to 90°C for 3 h. The samples were cooled to RT and extracted with 400 μl hexane. 250 μl of the organic phase were dried, the residue was re-suspended in 30 μl of pyridine and 30 μl of acetic anhydride and incubated for 1 h at 80°C. The samples were dried once again, resuspended in 60 μl of hexane, and immediately analyzed by GC-MS. The analytical method settings were identical to the targeted metabolomics method described above, with few modifications. The oven ramp was set from 80°C to 190°C at a rate of 5°C/min and then to 280°C at 15°C/min. D- and L-2HG peaks were quantified using the 173 m/z ion.
Brood size assay
Animals in the L4 larval stage were singled on 35 mm petri dishes. Every 24 h animals were moved to fresh plates until egg laying ceased. The remaining plates with embryos were incubated at 20°C for 24 h. Subsequently, L1 larvae and unhatched embryos were counted. Brood counts from animals that died or left the plate were excluded. For each biological replicate, data from at least seven animals were collected. The experiment was conducted three times.
Hatching assay
Approximately 30 synchronized L1 animals were placed on seeded 35 mm NGM agar plates. Animals were incubated at 20°C and allowed to lay eggs. Before eggs start hatching, adults were washed away and ∼300 embryos were transferred onto new plates. After 24 h of incubation, hatched larvae and unhatched embryos were counted to determine the rate of embryonic lethality.
Imaging
Differential interference contrast images were captured with a Zeiss Axioskop fitted with a Leica DFC360 FX camera. Confocal z-stacks were captured with a Leica TCS SP8 confocal microscope. Images were processed using ImageJ.
RNAi screen
RNAi clones of 2,104 C. elegans metabolic genes (30) were cultured in deep 96-well plates in LB (Miller) containing 50 μg/ml ampicillin and grown to stationary phase at 37°C. Cultures were concentrated 20-fold, and 15 μl were plated onto a shallow 96-well plate containing NGM agar supplemented with 64 nM vitamin B12 (adenosylcobalamin), 50 μg/ml ampicillin, and 1 mM IPTG. Plates were dried and stored overnight at 20°C. The next day 15 synchronized L1 animals were added to each well. Plates were screened for strong hatching defects on the 4th and 5th d of incubation at 20°C. The screen was performed three times. All hits were re-tested by performing a hatching assay on 35 mm NGM agar plates.
Statistical analysis
P-values were calculated using unpaired t test when comparing two conditions or Tukey’s test for multiple pairwise comparisons.
Acknowledgements
We thank Dr. Ralph DeBerardinis for advice on formate supplementation experiments. This work was funded by grants from the National Institutes of Health DK068429 to AJM Walhout and R35GM136315 to MV Sundaram.
Author Contributions
O Ponomarova: conceptualization, investigation, methodology, and writing – original draft, review, and editing.
AN Starbard: investigation and methodology.
A Belfi: investigation and methodology.
AV Anderson: investigation and methodology.
MV Sundaram: supervision, funding acquisition, investigation, and writing – original draft.
AJM Walhout: conceptualization, resources, supervision, funding acquisition, project administration, and writing – original draft, review, and editing.
Conflict of Interest Statement
The authors declare that they have no conflict of interest.
- Received July 2, 2024.
- Revision received July 4, 2024.
- Accepted July 4, 2024.
- © 2024 Ponomarova et al.
This article is available under a Creative Commons License (Attribution 4.0 International, as described at https://creativecommons.org/licenses/by/4.0/).