Commentary: Tumor suppressor p53 regulates heat shock factor 1 protein degradation in Huntington’s disease

Rocio Gomez-Pastor.

Department of Neuroscience, University of Minnesota, School of Medicine, Minneapolis, MN, United States.


Huntington’s disease (HD) is a rare, inherited neurodegenerative disorder marked by progressive motor impairment, cognitive decline, and psychiatric disturbances. Increasing evidence implicates dysregulated cellular stress responses in HD pathogenesis, particularly involving the antagonistic interplay between the transcription factors p53 and heat shock factor 1 (HSF1)—key regulators of both cancer biology and neurodegeneration. This commentary highlights recent findings by Mansky et al., which uncover a novel mechanism whereby mutant huntingtin (mtHTT) stabilizes p53, triggering the degradation of HSF1 in striatal neurons. This mechanistic link offers new insight into how disrupted proteostasis drives neuronal vulnerability in HD, highlighting the p53–HSF1 axis as a promising therapeutic target. Additionally, we examine evidence that similar regulatory dynamics may contribute to other neurodegenerative diseases, such as Parkinson’s and Alzheimer’s, suggesting a potentially conserved molecular pathway underlying neuronal degeneration in rare disorders. These findings underscore the therapeutic potential of modulating stress-responsive transcriptional networks to slow or prevent disease progression in HD and related conditions.


Huntington’s disease (HD) is a rare, inherited neurodegenerative disorder characterized by severe motor dysfunction, involuntary movements, cognitive impairment, and neuropsychiatric disturbances. The disease is caused by an expanded CAG trinucleotide repeat in the huntingtin (HTT) gene, resulting in a mutant huntingtin (mtHTT) protein with an extended polyglutamine (polyQ) tract 1. This pathogenic protein preferentially affects medium spiny neurons (MSNs) in the striatum—a brain region essential for motor control and aspects of cognition2. Accumulation of mtHTT disrupts a wide array of cellular processes, ultimately leading to neuronal dysfunction and death. In response to these disturbances, cells activate pathways of the integrated stress response (ISR) that rely on transcription factors to orchestrate survival and apoptotic programs. In neurons, the ability to mount effective stress responses is especially crucial, as their limited regenerative capacity renders them highly vulnerable to chronic proteotoxic and oxidative stress. Thus, understanding how stress-responsive transcriptional networks are altered in HD is essential for identifying potential targets to mitigate neurodegeneration. In the study by Mansky et al.3, the authors provide new mechanistic insight into how the tumor suppressor p53 destabilizes HSF1, a master regulator of protein homeostasis, in the context of HD. Their findings highlight a previously unrecognized crosstalk between these two transcription factors and suggest that the p53-HSF1 pathway may serve as a molecular target for the treatment of HD.

p53 is a well-known tumor suppressor that plays a crucial role in maintaining genomic stability by regulating cell cycle progression, apoptosis, and DNA repair4. In the context of HD, prior research has shown p53 levels are elevated in the striatum, especially in MSNs3,5. Interestingly, the elevated levels of p53 in patients with HD are associated with a reduced cancer incidence in this population6. p53 is highly regulated at the posttranslational level, particularly through protein degradation events, and there are several mechanisms in place to ensure appropriate levels of p53 in all cells, including MSNs7. In the study by Mansky and colleagues, the authors demonstrated that the accumulation of mtHTT enhances p53 stabilization by disrupting the interaction between p53 and its E3 ligase, MDM2, leading to increased p53 accumulation and transcriptional activity3. Among the different gene targets of p53, Protein Kinase CK2 alpha prime (CK2αʼ) and E3 ligase Fbxw7 are induced in MSNs. These two key proteins, act in a sequence of events to phosphorylate and ubiquitinylate, respectively, HSF1, promoting HSF1 proteosomal degradation (Figure 1).  HSF1 is a transcription factor canonically known for the regulation of protein quality control systems, including heat shock proteins, proteasomal proteins and proteins related to autophagy, which are all essential for maintaining protein homeostasis8. In addition, HSF1 also plays a role in the regulation of brain development, synapse stability, and memory consolidation9,10.

JRDRT-25-1219-fig1

Figure 1: Model for the stabilization of p53 and subsequent degradation of HSF1 in Huntington’s disease. Adapted from Mansky et al. 3

Previous studies have demonstrated that HSF1 levels are reduced in HD cell and mouse models and in affected brain tissues of HD patients, contributing to the accumulation of misfolded proteins and neuronal dysfunction11,12. The study by Mansky et al. provided a mechanistic connection between the previously reported reduced levels of HSF1 in HD and the upregulation of p53. These findings were further supported by experiments conducted in a conditional knockout of p53 in MSNs in the HD mouse model zQ175 that presented increased HSF1 protein levels, decreased HTT aggregation, and ameliorated motor deficits3

Interestingly, other studies have reported a complex crosstalk between p53 and HSF1 across various biological contexts. For example, research in human fibroblasts has shown that acute depletion of HSF1 induces cellular senescence through activation of the p53–p21 pathway13. Similarly, studies using mouse embryonic fibroblasts derived from Hsf1 knockout (Hsf1KO) mice have observed elevated p53 levels as a consequence of HSF1 deficiency 14. In that study, the authors demonstrated that loss of HSF1 leads to a reduction in αβ-crystallin, a molecular chaperone required for the recruitment of the E3 ubiquitin ligase Fbx4, which plays a role in p53 degradation. As a result, decreased αβ-crystallin levels lead to p53 stabilization in the absence of HSF1.

Moreover, HSF1 has been implicated in regulating p53 nuclear translocation and DNA-binding capacity in cancer cells 15,16. This bidirectional regulatory relationship is further supported by the observation that many cancer cells harboring p53 mutations, whether affecting its activity or stability, often exhibit elevated HSF1 expression17. Additionally, Hsf1KO mice have shown increased survival and reduced tumor incidence when exposed to chemical carcinogens and showed delayed tumor onset and decreased tumor burden compared to wild-type mice18. p53 has also been shown to inhibit HSF1 activity in certain tumor environments, although the mechanism here seems to be related to the activation of the CDKN1A/p21/MLK3 pathway19. Together, these findings highlight a dynamic and context-dependent interplay between HSF1 and p53, with potential implications for both tumor biology and neurodegeneration (Figure 2).

JRDRT-25-1219-fig2

Figure 2: Bidirectional regulatory relationship between p53 and HSF1.

The study by Mansky et al. contributes to shed light on the intricate molecular interactions between p53 and HSF1 in the context of neurodegeneration, especially in the pathogenesis of HD. By elucidating the mechanisms through which p53 regulates HSF1 degradation, the authors provided valuable insights that could inform the development of targeted therapies for HD. These findings suggest that modulating the p53-HSF1 pathway by reducing p53 levels and/or activity, or enhancing HSF1 stability may help restore protein homeostasis and mitigate neuronal dysfunction in HD. However, further research is needed to explore the potential side effects and efficacy of such interventions. While p53 inhibition shows promise as a neuroprotective strategy, especially in diseases like HD, the systemic risks, most notably cancer, cannot be overlooked. It is anticipated that systemic inhibition of p53 could lead to significant biological and therapeutic challenges due to the role p53 plays in regulating the cell cycle, DNA repair, apoptosis, and stress responses. Therefore, chronic or systemic p53 inhibition could lead to impaired DNA repair, unchecked cell proliferation, and genomic instability, greatly increasing the risk of cancer. The potential future of targeting p53 in HD likely lies in precise, targeted, and time-sensitive strategies that preserve the protective roles of inhibiting p53 while alleviating its neurotoxic consequences. A potential strategy would be to target local inhibition by using gene therapy, nanoparticles, or brain-region-specific promoters to inhibit p53 activity in striatal neurons, limiting systemic effects of p53.

On the other hand, strategies that enhance HSF1 activity and/or stability appear to be both safer and more promising. Pharmacological activation of HSF1 through inhibition of Hsp90, a central regulator of proteostasis and known regulator of HSF1, has been explored as a therapeutic approach in HD20. This strategy initially showed promise, as it enhanced the expression of HSF1 target genes and led to short-term improvement in disease phenotypes in HD mouse models. However, the therapeutic benefits were transient, and the approach was ultimately deemed ineffective for long-term disease modification. A key limitation of this approach lies in the fact that HSF1 levels are already markedly reduced in HD, restricting the pool of transcriptionally competent HSF1 that can be activated. However, it is worth noting that there is some discrepancy in the literature regarding HSF1 levels in HD models: while several studies report reduced HSF1 abundance in HD (9,11,12,21), others have found no significant decrease in certain mouse models22. Notably, even when HSF1 levels were preserved, HSF1 gene binding to target genes such as HSPs was still reduced and attributed to altered chromatin architecture that impeded a proper HSF1 activation of such targets23. Nevertheless, targeting Hsp90 carries the risk of widespread systemic disruption, which may outweigh the modest gains achieved by activating a limited reserve of HSF1.

A more promising approach involves the pharmacological inhibition of CK2α’, the rate-limiting enzyme responsible for HSF1 degradation in HD 11. CK2α’ is part of the CK2 holoenzyme, which is composed by two regulatory CK2β subunits and two catalytic subunits CK2αʼ and CK2α24. Both catalytic subunits share nearly 100% homology in their active site, but they differ in their abundance and substrate specificity. Contrary to CK2α, that phosphorylates hundreds of substrates and it is ubiquitously expressed throughout the body, CK2αʼ expression is more restricted to brain and testis and the number of substrates described for this subunit is more limited25. CK2αʼ is specifically elevated in HD MSNs, and systemic haploinsufficiency of CK2α’ in a mouse model of HD has shown to be safe, with no apparent side effects 11; 26. Moreover, this intervention restored HSF1 levels in the striatum, reduced HTT aggregation, improved striatal pathology, and rescued behavioral deficits 11; 26. Importantly, several CK2 inhibitors are commercially available and have received FDA orphan drug designation demonstrating they are safe for human use 27. However, a key limitation of these compounds is their lack of specificity, they inhibit both CK2α’ and its catalytic homolog CK2α28, which regulates hundreds of known substrates. Therefore, the use of non-subunit selective CK2 inhibitors could lead to several off-target effects. To address this, novel inhibitors specifically targeting CK2α’ have been recently developed, and although their use is still in early pre-clinical stages, have demonstrated efficacy in reducing HTT aggregation in HD cell models28. These targeted interventions aimed at increasing HSF1 levels represent a highly promising therapeutic strategy for the treatment of HD29.

Pathological degradation of HSF1 has also been observed in other neurodegenerative diseases. In Parkinson’s disease (PD), the E3 ligase NEDD4 promotes HSF1 degradation 30, while in Alzheimer’s disease (AD), this process is mediated by C/EBP-homologous protein (CHOP) and activation of the unfolded protein response (UPR) apoptotic pathway 31. Although no studies have explored whether CK2α’ is specifically altered in AD or PD, several studies using a pan antibody against CK2 (detecting both catalytic subunits simultaneously), or enzymatic activity assays, have reported elevated CK2 holoenzyme along with increased p53 levels32,33,34,35. However, whether p53 is involved in the elevation of CK2 in AD and PD and whether increased CK2 total level is due to a specific elevation of CK2α’ is still unknown. Elevated levels of FBXW7 have also been reported in neurons from parkKO mice, a model of hereditary PD 36 and there is evidence that FBXW7 (also known as SEL-10) may facilitate amyloid-β formation, implicating it in AD pathogenesis37. Although the specific downstream effectors of HSF1 degradation may differ between PD, AD, and HD, it remains plausible that a p53-dependent activation of CK2α’ and/or FBXW7 contributes to HSF1 destabilization across multiple neurodegenerative disorders (Figure 3). Investigating the p53–HSF1 axis in various neurodegenerative contexts may uncover a shared molecular mechanism underlying neuronal dysfunction and degeneration, while also offering insight into the broader relationship between neurodegeneration and cancer.

JRDRT-25-1219-fig3

Figure 3: Complementary mechanisms of HSF1 degradation across multiple neurodegenerative diseases. CK2 refers to the holoenzyme (CK2α’ + CK2α), while CK2α’ refers to the individual catalytic subunit. Upward black arrows indicate increased protein levels, while downward red arrows indicate reduction of protein levels.

In summary, the study by Mansky et al. provides compelling evidence that aberrant stabilization of p53 in HD contributes to the degradation of HSF1 via the CK2αʼ–FBXW7 axis, linking two critical regulatory proteins involved in cellular stress responses, proteostasis, and neuronal survival. These findings not only elucidate a novel molecular mechanism underlying HSF1 loss in HD but also open new avenues for therapeutic intervention aimed at restoring HSF1 function. Given HSF1’s role in maintaining protein quality control and neuronal integrity, strategies that stabilize HSF1, either through selective inhibition of CK2α′ or other targeted approaches, hold significant promise with potentially fewer systemic risks than direct p53 inhibition. Moreover, the broader implication that this p53–HSF1 regulatory pathway may operate in other neurodegenerative diseases such as PD and AD suggests a possible unifying mechanism driving neuronal vulnerability. Future research into brain-region-specific modulation of p53 activity and the development of CK2αʼ-selective inhibitors could pave the way for safer and more effective treatments not only for HD but for a range of neurodegenerative disorders, while also deepening our understanding of the complex interplay between neurodegeneration and cancer biology.

Conflict of interest

The authors declare no conflict of interest.

Funding

This work was funded by the University of Minnesota Medical School/UMF Bridge Award (R.G.P).

References

  1. MacDonald ME, Ambrose CM, Duyao MP, Myers RH, Lin C, Srinidhi L, et al. A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington's disease chromosomes. Cell. 1993;72(6):971–983. Available from: https://pubmed.ncbi.nlm.nih.gov/8458085/
  2. Vonsattel JPG, Difiglia M. Huntington disease. J Neuropathol Exp Neurol. 1998;57(5):369–384. Available from: https://doi.org/10.1097/00005072-199805000-00001
  3. Mansky RH, Greguske EA, Yu D, Zarate N, Intihar TA, Tsai W, Brown TG, Thayer MN, Kumar K, Gomez-Pastor R. Tumor suppressor p53 regulates heat shock factor 1 protein degradation in Huntington's disease. Cell Rep. 2023 Mar 7;42(3):112198. doi:10.1016/j.celrep.2023.112198. PMID: 36867535; PMCID: PMC10128052.
  4. Lanni C, Racchi M, Memo M, Govoni S, Uberti D. p53 at the crossroads between cancer and neurodegeneration. Free Radic Biol Med. 2012 May 1;52(9):1727–33. doi:10.1016/j.freeradbiomed.2012.02.034. PMID: 22387179.
  5. Bae BI, Xu H, Igarashi S, Fujimuro M, Agrawal N, Taya Y, et al. p53 mediates cellular dysfunction and behavioral abnormalities in Huntington's disease. Neuron. 2005;47(1):29–41. Available from: https://pubmed.ncbi.nlm.nih.gov/15996546/
  6. Sorensen SA, Fenger K, Olsen JH. Significantly lower incidence of cancer among patients with Huntington disease: An apoptotic effect of an expanded polyglutamine tract? Cancer. 1999;86(7):1342–6. Available from: https://www.ncbi.nlm.nih.gov/pubmed/10506723
  7. Momand J, Zambetti GP, Olson DC, George D, Levine AJ. The mdm-2 oncogene product forms a complex with the p53 protein and inhibits p53-mediated transactivation. Cell. 1992;69(7):1237–45. Available from: https://www.ncbi.nlm.nih.gov/pubmed/1535557
  8. Gomez-Pastor R, Burchfiel ET, Thiele DJ. Regulation of heat shock transcription factors and their roles in physiology and disease. Nat Rev Mol Cell Biol. 2018 Aug; Available from: https://www.ncbi.nlm.nih.gov/pubmed/28852220
  9. Zarate N, Intihar TA, Yu D, Sawyer J, Tsai W, Syed M, Carlson L, Gomez-Pastor R. Heat Shock Factor 1 Directly Regulates Postsynaptic Scaffolding PSD-95 in Aging and Huntington's Disease and Influences Striatal Synaptic Density. Mol Neurobiol. 2023 Apr;60(4):2035–49. doi:10.1007/s12035-022-03158-0. PMID: 36503102.
  10. Hooper PL, Tytell M, Vigh L, Musch MW, Kim-Colletti C. The central role of heat shock factor 1 in synaptic fidelity and memory consolidation. Cell Stress Chaperones. 2016;21(5):745-53. Available from: https://www.ncbi.nlm.nih.gov/pubmed/27283588
  11. Gomez-Pastor R, Burchfiel ET, Neef DW, Jaeger AM, Cabiscol E, McKinstry SU, et al. Abnormal degradation of the neuronal stress-protective transcription factor HSF1 in Huntington's disease. Nat Commun. 2017;8:14405. Available from: https://www.ncbi.nlm.nih.gov/pubmed/28194040
  12. Chafekar SM, Duennwald ML. Impaired heat shock response in cells expressing full-length polyglutamine-expanded huntingtin. PLoS One. 2012;7(5):e37929. Available from: https://www.ncbi.nlm.nih.gov/pubmed/22606270
  13. Oda T, Sekimoto T, Kurashima K, Fujimoto M, Nakai A, Yamashita T. Acute HSF1 depletion induces cellular senescence through the MDM2-p53-p21 pathway in human diploid fibroblasts. J Cell Sci. 2018 May 8;131(9):jcs210724. doi:10.1242/jcs.210724. PMID: 29632240.
  14. Jin X, Moskophidis D, Mivechi NF. Heat shock factor 1 deficiency via its downstream target gene alphaB-crystallin (Hspb5) impairs p53 degradation. J Cell Biochem. 2009;107(3):504–15. Available from: https://www.ncbi.nlm.nih.gov/pubmed/19343786
  15. Li Q, Feldman RA, Radhakrishnan VM, Carey S, Martinez JD. Hsf1 is required for the nuclear translocation of p53 tumor suppressor. Neoplasia. 2008 Oct;10(10):1138–45. doi:10.1593/neo.08430. PMID: 18813348; PMCID: PMC2546585.
  16. Logan IR, McNeill HV, Cook S, Lu X, Lunec J, Robson CN. Heat shock factor-1 modulates p53 activity in the transcriptional response to DNA damage. Nucleic Acids Res. 2009;37(9):2962-73. Available from: https://www.ncbi.nlm.nih.gov/pubmed/19295133
  17. Dai C, Sampson SB. HSF1: Guardian of proteostasis in cancer. Trends Cell Biol. 2016;26(1):17-28. Available from: https://www.ncbi.nlm.nih.gov/pubmed/26597576
  18. Dai C, Whitesell L, Rogers AB, Lindquist S. Heat shock factor 1 is a powerful multifaceted modifier of carcinogenesis. Cell. 2007;130(6):1005-18. Available from: https://www.ncbi.nlm.nih.gov/pubmed/17889646
  19. Isermann T, Breuer M, Aigner M, Schilling M, Lu T, Yu H, et al. Suppression of HSF1 activity by wildtype p53 creates a driving force for p53 loss-of-heterozygosity. Nat Commun. 2021;12(1):4019. Available from: https://www.ncbi.nlm.nih.gov/pubmed/34188043
  20. Riva L, De Michele G, Agostini F, Lin J, Tenore A, Sharma K, et al. Poly-glutamine expanded huntingtin dramatically alters the genome wide binding of HSF1. J Huntingtons Dis. 2012;1(1):33-45.
  21. Maheshwari M, Bhutani S, Das A, Mukherjee R, Sharma A, Katyal J, et al. Dexamethasone induces heat shock response and slows down disease progression in mouse and fly models of Huntington's disease. Hum Mol Genet. 2014;23(10):2737-51. Available from: https://www.ncbi.nlm.nih.gov/pubmed/24381308
  22. Gomez-Paredes C, Mason MA, Taxy BA, Papadopoulou AS, Paganetti P, Bates GP. The heat shock response, determined by QuantiGene multiplex, is impaired in HD mouse models and not caused by HSF1 reduction. Sci Rep. 2021 Apr 22;11(1):9117. doi:10.1038/s41598-021-88715-5. PMID: 33907289; PMCID: PMC8079691.
  23. Labbadia J, Cunliffe H, Weiss A, Katsyuba E, Sathasivam K, Seredenina T, et al. Altered chromatin architecture underlies progressive impairment of the heat shock response in mouse models of Huntington disease. J Clin Invest. 2011;121(8):3306-19. Available from: https://www.ncbi.nlm.nih.gov/pubmed/21785217
  24. Litchfield DW. Protein kinase CK2: structure, regulation and role in cellular decisions of life and death. Biochem J. 2003;369(Pt 1):1-15.
  25. Turowec JP, Duncan JS, Gloor GB, Litchfield DW. Protein kinase CK2 is a constitutively active enzyme that promotes cell survival: strategies to identify CK2 substrates and manipulate its activity in mammalian cells. Methods Enzymol. 2010;484:471-93.
  26. Yu D, Pait AM, Brenman JE, Dickey CA. CK2 alpha prime and alpha-synuclein pathogenic functional interaction mediates synaptic dysregulation in Huntington’s disease. Acta Neuropathol Commun. 2022;10(1):83. Available from: https://doi.org/10.1186/s40478-022-01379-8
  27. CX-4945 Granted Orphan Drug Designation. Oncology Times. 2017;23.
  28. Mudaliar D, Becker A, McKenna R, Petri S, Pagano N, Shioda T, et al. Discovery of a CK2α-Biased ATP-Competitive Inhibitor from a High-Throughput Screen of an Allosteric-Inhibitor-Like Compound Library. ACS Chem Neurosci. 2024;15(15):2703-18. Available from: https://www.ncbi.nlm.nih.gov/pubmed/38908003
  29. White A, McGlone A, Gomez-Pastor R. Protein kinase CK2 and its potential role as a therapeutic target in Huntington's disease. Biomedicines. 2022 Aug 15;10(8):1979. doi:10.3390/biomedicines10081979. PMID: 36009526; PMCID: PMC9406209.
  30. Kim E, Wang B, Sastry N, Masliah E, Nelson PT, Cai H, Liao FF. NEDD4-mediated HSF1 degradation underlies α-synucleinopathy. Hum Mol Genet. 2016 Jan 15;25(2):211–22. doi:10.1093/hmg/ddv445. PMID: 26503960; PMCID: PMC4706110.
  31. Kim E, Sakata K, Liao FF. Bidirectional interplay of HSF1 degradation and UPR activation promotes tau hyperphosphorylation. PLoS Genet. 2017;13(7):e1006849. Available from: https://www.ncbi.nlm.nih.gov/pubmed/28678786
  32. Alves da Costa C, Checler F, Vincent B. Presenilin-dependent gamma-secretase-mediated control of p53-associated cell death in Alzheimer's disease. J Neurosci. 2006;26(23):6377-85. Available from: https://www.ncbi.nlm.nih.gov/pubmed/16763046
  33. Herman AM, Liu MC, Sontag CJ, Kim M, Person MK, Kulesza RJ, et al. β-amyloid triggers ALS-associated TDP-43 pathology in AD models. Brain Res. 2011;1386:191-9. Available from: https://www.ncbi.nlm.nih.gov/pubmed/21376022
  34. Rosenberger AF, Morrema TH, Rozemuller AJ, van Haastert ES, Eikelenboom P, van der Flier WM, et al. Increased occurrence of protein kinase CK2 in astrocytes in Alzheimer's disease pathology. J Neuroinflammation. 2016;13(1):4.
  35. Takahashi M, Kanuka H, Fujiwara H, Kato N, Tanaka M. Oxidative stress-induced phosphorylation, degradation and aggregation of alpha-synuclein are linked to upregulated CK2 and cathepsin D. Eur J Neurosci. 2007;26(4):863-74. Available from: https://www.ncbi.nlm.nih.gov/pubmed/17714183
  36. Ekholm-Reed S, Hagerstrand H, Reed SI. Parkin-dependent degradation of the F-box protein Fbw7β promotes neuronal survival in response to oxidative stress by stabilizing Mcl-1. Mol Cell Biol. 2013;33(18):3627-43. Available from: https://www.ncbi.nlm.nih.gov/pubmed/23858059
  37. Li J, Zhou W, Fan J, Chen G, Zou H, Zhou Y, et al. SEL-10 interacts with presenilin 1, facilitates its ubiquitination, and alters A-beta peptide production. J Neurochem. 2002;82(6):1540-8. Available from: https://www.ncbi.nlm.nih.gov/pubmed/12354302
 

Article Info

Article Notes

  • Published on: June 30, 2025

Keywords

  • Huntington’s Disease
  • Neurodegenerative Diseases
  • p53
  • HSF1
  • Protein Homeostasis
  • Mutant HTT

*Correspondence:

Dr. Rocio Gomez-Pastor,
Associate Professor, Department of Neuroscience, University of Minnesota School of Medicine, USA;
Email: rgomezpa@umn.edu

Copyright: ©2025 Gomez-Pastor R.. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License.