A Study of the Effects of Phytohormones on Aging of Caenorhabditis Elegans

Palacký University in Olomouc Faculty of Science Laboratory of Growth Regulators & Department of Botany

A study of the effects of phytohormones on aging of Caenorhabditis elegans

Doctoral thesis

Mgr. Alena Kadlecová

Study Programme: P1527 / Biology Field of Study: 1507V004 / Botany Form: Full-time

Olomouc 2019 Supervisor: Prof. Ing. Miroslav Strnad, CSc. DSc. Consultant: Mgr. Jiríˇ Voller, Ph.D. Bibliographical identiﬁcation

Author’s first name and surname : Mgr. Alena Kadlecová Title of thesis : A study of the effects of phytohormones on aging of Caenorhabditis elegans Type of thesis : Doctoral Department : Laboratory of Growth Regulators Supervisor : Prof. Ing. Miroslav Strnad, CSc. DSc. Consultant : Mgr. Jiríˇ Voller, Ph.D. The year of presentation : 2019 Abstract : Cytokinins, a class of phytohormones involved in various aspects of plant growth and development, are also known to have diverse activities in animal models and humans. This thesis is directed to investigating the multiple protective effects of cytokinin bases which include anti-oxidative effect, anti-aging activity in skin cells and flies and neuroprotective effects in several in vitro and in vivo models. For detailed examination of these effects, we used the well-established model of aging – the nematode Caenorhabditis elegans. The results show that several cytokinins (both natural and synthetic derivatives prepared in our laboratory) – were able to increase the lifespan and/or stress resistance of the worms. We also carried out the initial steps to finding out more about the mechanism of action and metabolization of the best known active cytokinin – kinetin. Keywords : Cytokinins, kinetin, Caenorhabditis elegans, aging, longevity, stress resistance. Number of pages : 42 (+ 64 pages of supplements) Number of appendices : 4 Language : English Bibliografická identifikace

Jméno a príjmeníˇ : Mgr. Alena Kadlecová Název práce : Studium vlivu fytohormon˚una stárnutí Caenorhabditis elegans Typ práce : Dizertacníˇ Pracovišteˇ : Laboratorˇ r˚ustových regulátor˚u Vedoucí práce : Prof. Ing. Miroslav Strnad, CSc. DSc. Konzultant : Mgr. Jiríˇ Voller, Ph.D. Rok obhajoby práce : 2019 Abstrakt : Cytokininy, fytohormony podílející se na r˚ustua vývoji rostlin, jsou také známy pro své r˚uznorodéúcinkyˇ v živociších.ˇ V této práci jsme se zameˇriliˇ na protektivní efekt cy- tokininových bází. Ty jsou dle drívˇ ejšíchˇ studií schopny napríkladˇ zpomalovat stárnutí lidských kožních bunekˇ a octomilek a mají antioxidativní a neuroprotektivní úcinkyˇ v in vitro a in vivo modelech. Pro podrobnejšíˇ studium jejich aktivity jsme zvolili hád’átko obecné (Caenorhabdi- tis elegans), jakožto dobreˇ zavedený model stárnutí. Zjistili jsme, že radaˇ cytokinin˚u– a to jak prírodníchˇ látek, tak jejich syntetických derivát˚u– dokáže proloužit délku života a/nebo zvýšit rezistenci C. elegans ke stresu. Provedli jsme také pilotní exprerimenty s cílem lépe porozumetˇ metabolismu a mechanismu úcinkuˇ nejlépe prostudovaného aktivního cytokininu, kinetinu. Klícováˇ slova : Cytokininy, kinetin, Caenorhabditis elegans, stárnutí, dlouhovekost,ˇ odolnost v˚uciˇ stresu. Pocetˇ stran : 42 (+ 64 stran príloh)ˇ Pocetˇ prílohˇ : 4 Jazyk : Anglický I declare that this thesis is my original work and that I used only the sources listed in the Bibliography section.

In Olomouc, Mgr. Alena Kadlecová I would like to thank my supervisor Prof. Miroslav Strnad and my consultant Dr. Jiríˇ Voller for their input and advice. I would also like to thank our international collaborators, prof. Jan Kammenga from Wagengen University and Dr. Marta Artal-Sanz of the University Pablo de Olavide, as well as all of the members of their groups for allowing me to stay in their laboratories and learn a number of new valuable skills. I am also grateful to all of the members of Laboratory of Growth Regulators, especially the members of the "Anti-aging interventions" group for all of their support and advice, Dr. Ondrejˇ Novák and Hana Martínková for preforming the UHPLC- MS/MS analysis and Dr. Lenka Zahájská, Dr. Martin Hönig, Dr. Václav Mik and OlChemIm s.r.o for providing the test compounds. Last but not least, I would like to thank Mgr. Jan Michelfeit for his help with programming. Contents

1 Introduction and aims of this work 5

2 Literature review 6 2.1 Cytokinins ...... 6 2.1.1 Discovery of cytokinins ...... 7 2.1.2 Biosynthesis, metabolism and activity in plants ...... 7 2.1.3 Protective effect of cytokinins in animal models ...... 8 2.2 Caenorhabitis elegans ...... 10 2.2.1 C. elegans as a model of aging ...... 13

3 Materials and methods 15 3.1 Strains ...... 15 3.2 Compounds ...... 15 3.3 Basic cultivation protocols ...... 15 3.4 Chitinase assay ...... 16 3.5 Visual evaluation of lifespan in 96-well plates ...... 17 3.6 Stress assays ...... 17 3.6.1 Visual evaluation of stress resistance in 96-well plates ...... 17 3.6.2 Automated microscopy and image analysis ...... 17 3.7 Preparation of worm extracts and UHPLC-MS/MS analysis ...... 18

4 Results and discussion 19 4.1 Protective effect of natural cytokinin bases ...... 19 4.2 Protective effect of selected cytokinin derivatives ...... 26

5 Conclusions and future perspectives 29

References 30

6 List of published papers and other contributions 41

1 7 Supplements 43 7.1 Supplement 1 ...... 43 7.2 Supplement 2 ...... 56 7.3 Supplement 3 ...... 69 7.4 Supplement 4 ...... 82

2 List of abbreviations

C. elegans Caenorhabditis elegans cZ cis-zeatin

E. coli Escherichia coli tZ trans-zeatin

AMP adenosine monophosphate

AMPK AMP-activated protein kinase

ATP adenosine triphosphate

BAP 6-benzylaminopurine

DHZ dihydrozeatin

DMSO dimethylsulfoxid

FOXO Forkhead box protein O

FUDR ﬂuorodeoxyuridine

HTS high-throughput screening

IIS Insulin/insulin-like growth factor signalling pathway iPN 6-isopentenyladenine

K kinetin

KR kinetin riboside

KRDP kinetin riboside-5’-diphosphate

KRMP kinetin riboside-5’-monophosphate

3 KRTP kinetin riboside-5’-triphosphate

LGR Laboratory of Growth Regulators mT meta-topolin mTOR mechanistic target of rampamycin

NGM nematode growth medium oT ortho-topolin pT para-topolin

ROS reactive oxygen species

WT wild type

4 1 Introduction and aims of this work

Cytokinins are plant hormones that play a crucial role in plant growth and development. Various activities of cytokinins in animal models and humans have also been reported over the years. While many cytokinin ribosides possess anti-cancer activity, protective and anti-aging effects have been described for some cytokinin bases. The purpose of this work was to investigate the protective and pro-longevity effects of natural cytokinin bases and their derivatives using the well-established and convenient model organism, Caenorhabditis elegans (C. elegans). We studied the effect of natural cytokinin bases on the lifespan and stress resistance of the worms and performed initial experiments aimed at establishing the mechanism of action and understanding the metabolism of the best known active natural compound, kinetin. We also tested dozens of cytokinin derivatives from the unique chemical library of Laboratory of Growth Regulators (LGR), and found several that are potentially more active than their natural counterparts. An important part of this thesis was a literature survey that served as a basis of two reviews, in which we summarized the health-promoting activity of cytokinins, especially kinetin, in animal models.

5 2 Literature review

In this part of the thesis, we brieﬂy describe the discovery and activity of cytokinins in plants and animal models, as well as discuss the advantages of using C. elegans as a model of aging. For detailed information on the protective activity of cytokinins in animal models, the readers are referred to our published review article and book chapter ([1, 2], supplements 2 and 4).

2.1 Cytokinins

Cytokinins are a group of phytohormones implicated in various aspects of plant growth and development. Naturally occurring cytokinins are derivatives of adenine with either isoprenoid or aromatic side chain at position N6-. Examples of the best known cytokinin bases can be seen in Figure 1. Apart from bases, cytokinin ribosides, ribotides and conjugates with amino acids and sugars, most commonly glucose, can also be found in plants.

Figure 1: Structures of selected cytokinin bases. A) kinetin. B) N6-isopentenyladenine. C) 6-benzylaminopurine. D) para-topolin. E) meta-topolin. F) ortho-topolin. G) trans-zeatin. H) cis-zeatin. I) dihydrozeatin

6 2.1.1 Discovery of cytokinins

The discovery of cytokinins dates back to 1955, when Miller et al. isolated a derivative of adenine, N6-furfuryladenine, from autoclaved herring sperm [3]. In the presence of auxin, this compound was able to stimulate cytokinesis of plant cells and was therefore called kinetin (K). The first evidence of the natural occurrence of cytokinins came in the 1960s, when Letham and Miller partially isolated a compound which we now know as zeatin from maize (Zea mays) kernels [4]. K itself was long thought to be an artificial DNA rearrangement product. This changed in the 1990s, when K was found not only in plant extracts [5], but also in animal DNA [6] and later in human urine. [7]. Since then, various cytokinins were identified in other organisms apart from plants, including bacteria, fungi and mammals [8, 9, 10, 11].

2.1.2 Biosynthesis, metabolism and activity in plants

In plants, the biosynthesis of isoprenoid cytokinins is regulated by isopentenyl transferase [12, 13]. The enzyme converts ADP/ATP and dimethylallylpyrophosphate, produced by methylery- thritol phosphate pathway or the mevalonate pathway, into the active cytokinins, isopentenyladenosine-5’-diphosphate or -triphopshate. The isoprenoid side chain can be hydroxylated by cytochrome P450 enzymes, yielding trans-zeatin (tZ) ribotides. Ribotides can then be converted into their active, free base forms by the LONELY GUY family of enzymes [14]. That said, the biosynthesis of cytokinins with an aromatic side chain is not yet fully understood. On the other hand, cytokinin oxidases, enzymes able to cleave the N6- side chain, are responsible for the degradation of cytokinins [15, 13]. Plants can also decrease the levels of active cytokinins through their conjugation to glucose, calatyzed by glucosyl transferases. Glycosyla- tion can occur either on N3, N7 or N9 of the purine ring (N-glycosylation) or on the oxygen on the side chain of zeatins (O-glycosylation). Cytokinins are synthesized ubiquitously in plants but their long-distance transport is also possible both in xylem and phloem [16]. Passage through cell membranes is most likely facilitated by purine and/or nucleoside transporters [17]. After entering the cells, cytokinins act via binding to cytokinin receptors, the majority of which are localized in the endoplasmatic reticulum. The cytokinin signalling pathway is a His-Asp phosphorelay similar to the bacterial

7 two-component pathway, and involves the co-ordination of histidine kinase receptors, histidine phosphotransfer proteins and response regulators which perceive and relay the signal and mediate the response [13, 18, 19, 20]. Cytokinin signalling is indispensable in all stages of plant life, as it is involved in many growth and developmental processes. The biological effect of cytokinins is highly intertwined with the activity of other plant hormones, such as auxins [21] and ethylene. Cytokinins stimulate the division of a variety of plant cells and they regulate cell cycle progression, for example, via up-regulation of the expression of cyclin D3. On the other hand, they also promote differentiation, for example in root apical meristems and in the vascular system. Other functions of cytokinins in plants include: delaying leaf senescence, promoting chloroplast biogenesis, regulating the circadian clock, inﬂuencing nutrient uptake and modulating response to abiotic stress. [13, 22]

2.1.3 Protective effect of cytokinins in animal models

Remarkably, cytokinins also have a variety of activities in animal models. Whereas cytokinin ribosides are well known for their cytotoxic and anti-cancer effects [23], cytokinin bases are reported to have protective activity which we have summarized in detail here ([1, 2], supplements 2 and 4). Briefly, the most studied cytokinin in this regard is kinetin, which was reported to have anti-aging effects on cultured skin fibroblasts [24], invertebrates [25, 26, 27] and, mammalian and human skin [28, 29, 30, 31, 32, 33]. tZ was also able to modulate aging in cultured skin fibroblasts [34], and also found to protect skin cells against photodamage and improve their healing/migration capability [35, 36]. Both compounds are currently used in cosmetics. The success of cytokinins also led to the development of synthetic derivatives for use in dermatology, the most notable example of which is 6-furfurylamino-9-(tetrahydropyran-2-yl)-9H-purine (Pyratine) [37, 38, 39]. Interestingly, K is also able to modulate the differentiation of skin keratinocytes [40, 41], including cells derived from psoriatic patients [42]. Morevover, K is reported to have protective activity in several models of central nervous system stress. It prevented glutamate-induced oxidative damage in a mouse hippocampal cell line [43], protected primary culture of rat astrocytes against damage induced by the administration

8 of D-galactose [44]. It also showed (neuro)protective activity in rodents subjected to radiation damage [45] and co-treatment with D-galactose and aluminum [46]. Neuroprotective activity was also described for tZ. It protected the rat pheochromocytoma cell line against β-amyloid induced toxicity, as well as improving cognitive function in mice with scopolamine-induced amnesia [47, 48]. In the past few years, K has also emerged as an interesting candidate for the treatment of some forms of hereditary neurodegenerative disorders, such as early onset Parkinson’s disease associated with mutations in the enzyme PINK-1 [49] and Huntington’s disease [50]. The authors of these studies reported that K can be transformed by cells into kinetin riboside-5’- monophosphate and -triphosphate (KRMP, KRTP, Figure 2) and that the latter acts as a neo- substrate of disease relevant kinases, PINK-1 and casein kinase 2, respectively, boosting their activity.

Figure 2: Structures of A) Kinetin riboside. B) Kinetin riboside-5’-monophosphate. C) Kinetin riboside-5’- triphosphate.

K is also able to correct aberrant pre-mRNA splicing of ELP-1 gene [51], mutations of which cause familial dysautonomia (FD) – a rare but debilitating hereditary disease characterized by impaired development and survival of neurons in both sensory and autonomic nervous systems [52]. Remarkably, K was able to increase the amount of correctly spliced product not only in various types of in vitro cultured cells [51, 53, 54, 55, 56, 57], but also in rodents [58] and human volunteers [59, 60]. According to ClinicalTrials.gov, it is currently in phase II of clinical trials (Identiﬁer: NCT02274051). A similar effect of K was also observed in models of another splicing disorder, neuroﬁbromatosis I [53, 61], although it must be noted that due to the fact that many different mutations cause this disease (unlike FD, where a single mutation is responsible for more than 99 % of cases), the applicability of K may be limited. In several studies, K was able to delay the physiological decline and reduce markers of

9 oxidative damage in various tissues of rodent models of accelerated aging [62, 46, 63, 64, 65]. There is also some evidence that K may improve some aspects of cardiovascular [66, 67, 68] and metabolic [69] diseases. The mechanism of action of cytokinins is not yet fully understood. Their effect is often ascribed to their ability to reduce oxidative stress, both by directly acting as a ROS scavenger [70, 71, 72, 73, 67] and by inducing the cell’s own defence mechanisms [43, 26]. K was also proposed to act as a hormetin [74, 75]. Hormesis can be deﬁned as an adaptive response to mild stress, resulting in activation of a cell’s own protective mechanisms, leading to a beneﬁcial effect. The agents that cause such response are called hormetins [76, 75]. As outlined above, K is also able to modulate alternative splicing of pre-mRNA [51], although it is not yet clear how, and its metabolite, KRTP, acts as a neo-substrate of at least two disease-relevant kinases [49, 50]. It is not yet known whether these various activities are connected and if the effect reported in different cells and organisms – often similar in nature – is a result of a shared mechanism of action.

2.2 Caenorhabitis elegans

C. elegans is a free-living soil nematode which can typically be found in temperate regions in rotting plant material, where it feeds on bacteria [77]. In the 1960s, Sydney Brenner proposed to use the nematode as a simple and convenient model organism for studying developmental biology and neurobiology [78]. Its success and many advantageous characteristics soon resulted in adapting C. elegans in other biological fields. Today, C. elegans is used in hundreds of labs worldwide and it has facilitated a number of breakthrough discoveries. As an example, let us name three that resulted in the Nobel prize – characterization of the genetic regulation of organ development and apoptosis (NP in physiology and medicine for Brenner, Sulston and Horovitz in 2002) [79, 80], the discovery of RNA interference (NP in physiology and medicine for Fire and Mello, 2006) [81] and using green fluorescent protein as a marker of gene expression [82] (NP in chemistry, for, among others, Chalfie who used C. elegans, 2008). The worm has many advantageous characteristics that have promoted its popularity. It is safe, cheap and easy to maintain in a lab. C. elegans is a small animal, with adults measuring

10 only about 1 mm in length and 80 µm in diameter. It can be cultivated on agar plates seeded with bacteria, easily age-synchronized by sodium hypochlorite treatment thanks to the highly resistant, chitin-containing egg shells and after transferring into glycerol-based freezing solution, it can survive storage in liquid nitrogen and thawing. The worm is transparent, which enables researchers to observe changes at a single cell level under a microscope, to use ﬂuorescent markers etc. C. elegans has a rapid life cycle – in 20 ◦C and under favourable conditions, it goes from egg through 4 larval stages (L1 - L4) to egg laying adults in just 3 days (Figure 3, A). Adults are capable of producing hundreds of progeny within a week. When food is sparse and the population density too high, C. elegans can form an alternative L3 stage called a dauer. These larvae do not feed or reproduce, and are long-lived and highly resistant to various types of stresses. When the outer conditions become favourable, dauers resume development and grow into standard adults [77]. Moreover, the tight regulation of the development, results in a constant number of cells and nearly invariable development of all animals, allowing researchers to map the lineage of every cell in the body with extraordinary precision and study the genes involved in the development. There are two sexes of C. elegans adults – hermaphrodites (XX) capable of self-fertilization and males (X0) (Figure 3, B). The males are anatomically distinct from the hermpahrodites, and they comprise only approximately 0.1 % of population. They appear either if a non-disjunction of sex chromosome occur or after mating of a hermaphrodite with a male, which results in higher number of progeny (up to approx. 1000 compared to approx. 300 after self-fertilization), half of which are males. The existence of the males as well as the possibility of self-fertilization is one of the reasons why C. elegans is a very powerful tool in genetics – if an interesting mutation occurs during mutagenesis, it can stabilize in a population by the self-propagation of hermaphrodites. At the same time, the possibility of crossing enables introduction of new mutations into a speciﬁc genetic background etc. Apart from creating mutants, the functions and interactions of genes can also be studied by RNA interference (RNAi) experiments. In C. elegans, these experiments can be performed very easily – the double-stranded RNA can be introduced to the worms by simply feeding them transformed bacteria. All this makes the worm exceptionally convenient for both reverse and forward genetic screening.

11 Figure 3: Overview of life cycle of C. elegans in 20 ◦C. Credit: Ann K. Corsi, Bruce Wightman, and Mar- tin Chalﬁe (2015), A Transparent window into biology: A primer on Caenorhabditis elegans, WormBook (http://www.wormbook.org/)

Apart from sex chromosomes, C. elegans has 5 pairs of autosomes. Its genome has several interesting features – such as the fact that the chromosomes are holocentric and the existence of genes organized in operons [83]. The genome is rather compact, containing only 100 000 000 base pairs. Unsurprisingly, the worm was the ﬁrst multicellular organism whose genome was fully sequenced [84]. C. elegans is a eukaryote, therefore a signiﬁcant percentage of its genes are homologous to mammalian ones. These include a number of disease-relevant genes in humans [85]. The same is true for many molecular pathways that are conserved through the phylla. Due to these similarities, C. elegans is frequently used to study genetic and molecular aspects of various human diseases and can also be used for drug screenings [77, 86]. The small size, transparency and possibility of cultivation in liquid-based media are other signal advantages of the worm which is especially relevant for drug screening – it enables whole- organism high-throughput approaches [87].

12 2.2.1 C. elegans as a model of aging

Aging can be characterized as a progressive decline in physiological processes leading to impaired functioning of the organism and increased risk of disease and, ultimately, death [88, 89]. Although the molecular mechanisms underlying aging are not yet fully understood, consider- able advances have been made in aging research the past few decades and many of them have been facilitated by C. elegans. In the 1970’s, C. elegans was proposed as a suitable model for studying aging [90, 91]. Apart from all of the advantages mentioned above, the worm has a short lifespan of only about 3 weeks on average, making aging experiments less time- and resource-intensive than, for example, using mammalian models. Aging worms undergo quantifiable changes equivalent to human aging, such as decrease in movement, sarcopenia, decline in sensory perception, reduced metabolic rate and accumulation of pigment lipofuscin. Understandably, however, using C. elegans as a model of human aging has its disadvantages. Most notably, the worm has a very simple body plan and lacks many tissue types and organs [91]. However, C. elegans is a well differentiated organism with many cell types similar to mammalian cells [92]. A few years after the worm was first proposed as a suitable model of aging, its usefulness was supported by discovery of the first genes able to significantly prolong the lifespan [93, 94]. Long-lived animals had mutations in components of the insulin/insulin-like growth factor signalling pathway (IIS) – a conserved pathway responsible for nutrient sensing that subsequently regulates the organism’s growth, development and metabolic rate. IIS was the first signalling pathway associated with aging and longevity and, its diminished signalling affects the lifespan of multiple organisms, from yeast to possibly humans [95]. Since then, multiple other conserved proteins and pathways associated with the longevity of various species have been discovered, and C. elegans plays a significant role in unravel- ling their functions and interactions [96]. These proteins/pathways are commonly involved in metabolism and nutrient sensing, mitochondrial function and stress resistance, and their functioning is tightly interconnected. Among these belong for example sirtuins, NAD-dependent protein/histon deacetylases, which, among others, regulate transcription, stress response, apop-

13 tosis and mitochondrial biogenesis [97, 98]; mechanistic target of rampamycin (mTOR), a ser- ine/threonine kinase that controls cell growth and metabolism in response growth factors, nutrients, the energy status of the cell and stress [99]; AMP-activated protein kinase (AMPK), which provides information about cell energy levels by sensing ATP/ADP/AMP ratios and subsequently regulates metabolism [100]; and others [101, 102]. Apart from understanding the genetics of aging and interactions between longevity-associated genes, C. elegans is also used for routine screening of compounds with the potential to affect the lifespan. Such compounds would have potential use medicine and provide relief for elderly people suffering from age-related pathologies, such as neurodegenerative, cardiovascular and metabolic diseases. As outlined in this part of the thesis and here ([1, 2], supplements 2 and 4), cytokinins could be one group of such compounds, and for studying their effect, we chose to use C. elegans as a convenient model.

14 3 Materials and methods

3.1 Strains

The bacterial strain used in this study was Escherichia coli OP50 (E. coli). C. elegans strains used in this study were: N2 (wild-type Bristol), BA17 (fem-1) and CF1038 (daf-16(mu86)). All strains were purchased from Caenorhabditis genetic center (http://cbs.umn.edu/cgc/home).

3.2 Compounds

Natural cytokinin bases used in this study – kinetin (K) , 6-benzylaminopurine (BAP), ortho- topolin (oT), meta-topolin (mT), para-topolin (pT), N6-isopentenyladenine (iP), trans-zeatin (tZ), dihydrozeatin (DHZ), cis-zeatin (cZ) – were kindly provided by Olchemim s.r.o. Syn- thetic derivatives tested as a part of this thesis were prepared and provided by colleagues from Laboratory of Growth Regulators (LGR), Lenka Zahájská, Martin Hönig and Václav Mik. Their preparation was described in detail here [103, 104] (supplement 3). Prior to the testing, all compounds were dissolved in dimethylsulfoxide (DMSO) to achieve 100 mM stock solutions. These were then added to cultivation media and therefore further diluted to reach the desired concentrations (up to 200 µM for the natural cytokinins, up to 100 µM for the derivatives).

3.3 Basic cultivation protocols

Worms were cultivated using standard protocols, buffers and media described in a peer-reviewed online resource WormBook (http://www.wormbook.org/, [105]). Maintenance cultures were kept in 15◦C. Experimental cultures were cultivated in 20◦C, or, in case of BA17 (fem-1), in 25◦C. At this temperature, the mutants develop into females and are unable to reproduce. To avoid contamination with progeny in experiments in which other strains were used, 25 µg/ml of ﬂuorodeoxyuridine (FUDR) was added to the media. Worms were fed with uracil auxotroph strain of E. coli, OP50. Using this strain ensures that the bacteria do not overgrow in uracil poor media used for cultivation of C. elegans. Bac-

15 teria were cultivated overnight in 37◦C in LB medium. For solid media cultivation of worms, bacteria were streaked onto NGM (nematode growth medium) plates, spread and cultivated either overnight in 37◦C or for 2 days at room temperature. For liquid media cultivation of worms, bacterial suspension in LB was transferred into pre-weighted centrifugation tubes and centrifuged (3000 x g, 15 minutes). Pellets were washed twice with sterile distilled water. The wet pellets were then weighed and re-suspended in S medium. Age-synchronized populations were prepared by bleaching. Populations containing preg- nant hermaphrodites were washed from plates using M9 buffer into microcentrifuge tubes and brieﬂy centrifuged. The excess liquid was removed and 1 ml bleaching solution (for 10 ml: 8 ml water; 1,5 ml sodium hypochlorite 14% of chlorine in aqueous solution; 0,5 ml 10 M NaOH) was added to the worm pellet. The suspension was placed on a thermoshaker (1500 rpm) for 5 to 10 minutes. The pellets were washed 3 times with M9 and twice more with S complete medium in case experiments with liquid cultures were planned.

3.4 Chitinase assay

The short-term toxicity of natural cytokinin bases to the larvae and their effect on worm fecundity was measured via a chitinase assay [106] as well as by visual evaluation of the worms under a microscope. Briefly, freshly hatched synchronized wild type (WT) L1 larvae were diluted to a final concentration of 200 - 300 worms per ml, fed with 3 mg/ml bacteria and distributed onto 96-well plates. Larvae were then treated with cytokinins or vehicle alone (DMSO). Populations were left to grow at 20◦C while shaking on a shaker (100 rpm) for 4 days. In this time, the worms would reach adulthood and start reproducing. 20 µM fluorogenic chitinase substrate (4-methylumbelliferyl-β-D-N,N0,N00-triacetylchitotrioside) was added to each well and experimental plates were incubated at 37◦C for 1h. The reaction was then stopped by the addition of alkaline buffer (1 M glycine/1 M NaOH, pH 10.6), and the intensity of the fluorescence was measured on a plate reader. Prior to the assay, the plates were visually checked under a microscope for abnormalities, such as dead worms or developmental impairments.

16 3.5 Visual evaluation of lifespan in 96-well plates

The lifespan of experimental populations cultivated in liquid medium in 96-well plates was measured by manual counting of living worms in an inverted microscope [107]. Brieﬂy, age synchronized BA17 young adults were fed with 6 mg/ml of bacteria, treated with cytokinins and transferred onto 96-well plates. Plates were carefully sealed and the seal was removed once a week to ensure the worms had enough oxygen. The number of surviving individuals was counted every 3 days. The method, as well as statistical tools and software used for the analysis of the results, are described in detail here [27] (supplement 1).

3.6 Stress assays

3.6.1 Visual evaluation of stress resistance in 96-well plates

For part of the study, the resistance of experimental populations cultivated in liquid medium in 96-well plates to oxidative/heat stress was measured by visual counting of surviving worms in an inverted microscope. Brieﬂy, age synchronized young adults were fed with 6 mg/ml of bacteria, treated with cytokinins and transferred onto multi-well plates or small Petri dishes. After 3-days cultivation, which would be sufﬁcient for compounds to induce a response, the animals were washed and transferred onto 96-well plates that are more suitable for counting. The experimental populations were then subjected to to either heat (35◦C) or oxidative (juglon 500 µM) stress and surviving worms were counted under a microscope. The method, as well as statistical tools and software used for the analysis of results, are described in detail here [27, 104] (supplements 1 and 3).

3.6.2 Automated microscopy and image analysis

Some experiments were performed using automatic microscopy and image analysis. This process required specialized equipment not available in the LGR at the moment, which is why these experiments were conducted during 3-months internship in Marta Artal Sanz’s group in the Andalusian Centre for Developmental Biology, Universidad Pablo de Olavide, Sevilla. For microscopy, image analysis and parts of the high-throughput screening, we used protocols pre-

17 viously established in the hosting laboratory and published in Hernando-Rodríguez et al., 2018 [108]. For stress assays, we used conditions previously optimized in LGR. The overview of the protocol can be seen in Figure 4. Results were analyzed using scripts in the programming language Python (available at https://github.com/kadlecova-alena/Stress-scripts), that were kindly created for us by Mgr. Jan Michelfeit. The method was used for preliminary evaluation of the protective activity of cytokinin derivatives.

Figure 4: Overview of the protocol of high-throughput stress assays.

3.7 Preparation of worm extracts and UHPLC-MS/MS analysis

Adult worms and bacteria were cultivated with 100 µM K for 24h. The biological material was then washed 4 times, snap frozen with liquid nitrogen and stored in -80◦C. The precise preparation and UHPLC-MS/MS measurement of the levels of kinetin and its metabolites was performed by Hana Martínková and Dr. Ondrejˇ Novák, and described in detail here [27] (supplement 1).

18 4 Results and discussion

In the experimental part of this thesis, we focused on investigating the pro-longevity and stress resistance-inducing effect of both natural cytokinin bases and their derivatives prepared in LGR. In this section, we present and discuss the main ﬁndings of our project. The detailed results and further discussion can be found in the publications [27, 1, 104] (supplements 1,2,3).

4.1 Protective effect of natural cytokinin bases

First, we tested the pro-longevity effect of 9 natural cytokinin bases (for details see [27], supplement 1). The set consisted of 5 cytokinins with an aromatic side chain (kinetin (K), 6- benzylaminopurine (BAP), ortho-topolin (oT), meta-topolin (mT) and para-topolin (pT)) and 4 cytokinins with isporenoid substitution (N6-isopentenyladenine (iP), trans-zeatin (tZ), dihydrozeatin (DHZ) and cis-zeatin (cZ)). None of the compounds had any acute toxic effect in C. elegans, nor did they signiﬁcantly reduce fecundity in the chitinase assay (Figure 5).

Figure 5: Natural cytokinin bases do not inﬂuence short-time survival or affect the fecundity of the worms. Error bars show SD in 3 repeated experiments.

In the initial screening, we identified four compounds that were able to significantly prolong the lifespan of the worms. Three of them were aromatic cytokinins – K, pT and mT (Figure 6). For the remaining active cytokinin, tZ, however, we could not confirm its effect in repeated experiments with bigger populations.

19 Figure 6: Three natural cytokinin bases signiﬁcantly (p < 0.05, Log-rank test) and repeatedly prolonged the lifespan of C. elegans. A) Kaplan-Meier survival curve of population treated with 200 µM kinetin compared to control. B) The Kaplan-Meier survival curve of populations treated with 50 µM meta-topolin compared to control. C) Kaplan-Meier survival curve of population treated with 100 µM para-topolin compared to control.

Para-topolin was reported to have antioxidant activity in vitro [109], and a recent study also showed that its topical application improves signs of photodamage in human skin [110]. Other than that, not much is known about the (cyto)protective activity of topolins in cell cultures and animals so far. Our results suggest that their further investigation might be worthwhile. Nevertheless, K is the most studied cytokinin with protective effects in animal models to date. This is why we selected it as a representative compound for follow-up experiments, as the data we obtained could be placed in the wider context of previously published results.

20 Commonly, interventions that prolong the lifespan of worms also increase their stress resistance [111]. In accordance with this trend, 3 day pre-treatment with K, increased worms’ thermotolerance and resistance to superoxide anion generator [112] juglone (Figure 7).

Figure 7: Kinetin increases the survival of C. elegans under stress. A)The survival of worms after heat stress (35◦C, 90 minutes). Error bars show the standard deviation in three repeated experiments and asterisk indicates statistical signiﬁcance (p < 0.05, the Z test for population proportions) in each experiment. B) Kinetin 200 µM signiﬁcantly (p < 0.05, Log-rank test) and repeatedly increased survival of worms after exposure to ROS generator (juglone 500 µM).

Next, we decided to perform initial steps to study the mechanism of action of K in C. elegans. The beneﬁcial effect of K we observed could be explained by its interaction with one of the central longevity-associated pathways, insulin/insulin-like growth factor pathway (IIS). The main effector of this pathway is the transcription factor DAF-16/FOXO (Forkhead box protein O). Decreased IIS signalling results in translocation of DAF-16 into the nucleus and trans- activation of its target genes. As daf-16 is downstream of other components of the pathway, such as well known pro-longevity genes daf-2 and age-1, it is required for their beneﬁcial effect [113]. This underlay our decision to test the effect of K in daf-16(mu86) mutants with dysfunctional DAF-16 and we observed that K increased the stress resistance in these mutants as well (Figure 8). It therefore seems that K’s effect is not DAF-16 dependent.

21 Figure 8: Kinetin increases survival of daf-16 mutants under stress. A) Survival after heat stress (35◦C, 90 minutes). Error bars show the standard deviation in three repeated experiments and asterisks indicate statistical sig- niﬁcance (p < 0.05, the Z test for population proportions) in each experiment. B) Kinetin signiﬁcantly (p < 0.05, Log-rank test) and repeatedly increased worm survival after exposure to ROS generator (juglone 500 µM).

The protective effect of cytokinins in animal models was previously ascribed to their ability to protect against oxidative stress. Reactive oxygen species (ROS) play an important role in aging. Traditionally, ROS were considered to be a crucial pro-aging factor because they cause direct damage to cell macromolecules (free radical theory of aging [114]). Today, we know that their role is much more complex – they can for example act as signaling molecules, directly modify multiple proteins involved in signal transduction and they can also influence epigenetic modulation of gene expression [115]. Apart from this, ROS can also activate adaptive responses – this process is also known as mitochondrial hormesis or mitohormesis [116]. As K was previously proposed to act as a hormetin, we decided to test how the presence of an antioxidant – analogue of vitamin E, Trolox – would influence the effect of K. Notably, in these co-treatment experiments the effect of K was significantly diminished in both stress assays and lifespan experiments (Figure 9). These results suggest that the mechanism of K’s action in worms cannot be ascribed to simply acting as a direct antioxidant – on the contrary, it seems that the presence of ROS is required for the beneficial effect of K. Our results are consistent with the hypothesis that K acts as a hormetin.

22 Figure 9: The beneficial effect of kinetin is no longer apparent if the worms are co-treated with an antioxidant. A) Survival after heat stress (35◦C, 90 minutes). Error bars show the standard deviation in three repeated experiments and asterisks indicate statistical significance (p < 0.05, Z test for population proportions) in each experiment. B) Survival after oxidative stress (juglone 500 µM). Kinetin significantly (p < 0.05, Log-rank test) increased survival of worms compared to kinetin + trolox co-treated worms after exposure to ROS generator. C) Kaplan-Meier survival curve of a population treated with kinetin alone compared to population co-treated with kinetin and trolox and control. Kinetin significantly (p < 0.05, Log-rank test) increased longevity of worms compared to kinetin + trolox co-treated worms.

Further, we conducted experiments aimed to determine how efﬁciently is K absorbed and how is it metabolized. We measured the levels of K in extracts prepared from non-treated and K-treated worms, and to assess how the bacteria used as food inﬂuence the metabolism, we also tested extracts from the bacteria and worms fed with heat-killed bacteria.

23 sample type kinetin kinetin riboside kinetin riboside-5’-monophosphate C. elegans, standard cultivation conditions 160.34 17.58 372.59 970.14 71.5 508.23 C. elegans, heat killed bacteria 33.83 9.47 30.51 80.55 195.15 127.61 E. coli OP50 0.54 0.31 ND 0.28 0.22 ND

Table 1: Concentrations (µM) of kinetin and its metabolites in extracts from worms and bacteria after treatment (100 µM kinetin, 24 hours).

No K was detected in samples from non-treated worms and bacteria. This suggests that, at least under the cultivation conditions we used, K is not naturally formed in these organisms. This finding is rather unexpected, as K was previously proposed to be formed in a variety of different organisms as a result of DNA damage [117, 118, 119]. The published data concerning the natural occurrence of K is currently rather conflicted, as it was detected in native samples in some studies [6, 7, 58, 120], but not in others [121]. We also did not detect K in plant and cell extracts in some of our own experiments (unpublished). These results prompted us to conduct a follow-up study which is currently underway, in which the presence and metabolization of K and other cytokinins will be investigated in various model organisms and mammalian cells. Samples from cells/worms subjected to various types of stress will also be tested to see whether K occurs at detectable levels, which would provide support for the currently accepted theory that K is formed during DNA damage. Exogenously applied K was efficiently absorbed by the worms and was also found in bacteria (Table 1). Apart from K, we also detected kinetin riboside (KR) and kinetin riboside-5’- monophosphate (KRMP). The analytic method we used, did not allow us to detect the higher phosphates. It is therefore unclear whether KRDP and KRTP are also formed in the worms. All 3 metabolites were found in extracts from worms fed with heat-killed bacteria, showing that C. elegans can effectively ribosylate K. In bacteria, KR but not KRMP was detected. The presence of these metabolites presents an interesting possibility that K might act as pro-hormetin and it is actually its metabolites that cause the hormetic response. KR is well

24 known cytotoxic compound able to induce ATP depletion [122, 123], possibly by impairing mitochondrial function [124]. Low doses of KR might therefore trigger an adaptive response in C. elegans. Moreover, low doses of some cytokinin ribosides were previously reported to activate nuclear factor erythroid-derived 2-like 2 (Nrf2) [125] – a transcription factor playing an important role in regulating the stress response [126] – and some of our previously unpublished data suggest KR might have the same effect as well. Interestingly, K too was shown to activate Nrf2 in a mouse hipoccampal cell line [43]. Apart from the presence of KR and KRMP, it also worthwhile to consider that metabolites such as α,β-unsaturated dialdehydes, arising from metabolisation of the furan ring, might act as hormetins and/or Nrf2 activators [127]. Taking our results together, there might be one more interesting possible explanation of K’s beneﬁcial effect – K or its metabolites might act as an activator of AMPK. Both K and KR were previously reported to activate AMPK in cells [128] and one other cytokinin, N6- isopentenyladenosine-5’monophosphate, was proposed to activate AMPK by binding to its γ- subunit instead of AMP [129]. It is tempting to speculate that KRMP, into which K/KR can be transformed in worms and cells [130], can do that as well. It is known that the effect of AMPK can be suppressed by addition of antioxidants [131, 132, 133], as we saw in the case of K. The effect of AMPK on worm aging is also likely to be at least partly independent of DAF-16, as the C. elegans daf-16 and aak-2 (gene coding AMPK in worms) double mutants have shorter lifespan than either single mutant [134]. Moreover, metformin, an antidiabetic drug able to prolong the lifespan of multiple species most likely via AMPK activation [135], also prolongs the lifespan of daf-16 mutants [136]. In summary, our results provide the basis for several hypotheses of how K might induce the longevity and stress resistance in C. elegans. More experimental evidence is needed to draw any deﬁnite conclusions. This is why we also aim to conduct a follow-up study, in which we will investigate whether SKN-1 (a C. elegans homolog of Nrf2) is activated in K-treated worms by using reporter strains and real-time PCR, as well as assess whether treatment with K leads to activation of AMPK in worms and cells using Western blotting and by testing the effect of K in C. elegans strains with mutated aak-2.

25 4.2 Protective effect of selected cytokinin derivatives

The protective activity of natural cytokinins prompted us to investigate their synthetic derivatives. Over the more than 20 years of existence of LGR, hundreds of cytokinin derivatives were prepared for application in medicine, cosmetics and agriculture, resulting in a unique chemical library. One notable example of cytokinin derivative with protective activity created by LGR is 6-furfurylamino-9-(tetrahydropyran-2-yl)-9H-purine, a derivative of K which is currently used in cosmetics and marketed under the trade name Pyratine. As we outlined above, interventions that lead to prolonged lifespan also often result in increase in stress resistance [111], which was also true for K. This close link between the two activities proves to be useful for drug screenings for compounds modulating aging, as stress assays are more suitable for high-throughput screening (HTS) methods. This is why we, in cooperation with Dr. Marta Artal-Sanz’s group from Universidad Pablo de Olavide, decided to conduct HTS for compounds increasing resistance of C. elegans to oxidative and heat stress. Our test set included derivatives of natural cytokinin bases with substitutions at various positions of the purine ring. The tested set was enriched for compounds with protective activity, with 24 compounds out of 73 being able to signiﬁcantly increase resistance to one of both types of stresses in the initial screening (Z-test for population proportions). The majority of the hits have a substitution at position 8 of the purine ring (Figure 10) and some of them also possess a 9-(tetrahydropyran-2-yl) group, the same as the approved cosmaceutical Pyratine. The preparation of these compounds is described in detail here [103]. It is also important to note that natural cytokinins were also included in the set. Notably, many of the active compounds were more active than their natural counterparts, partly maybe because here, the compounds were tested at a lower concentration than in the study on the protective activity of natural cytokinin bases – for example, K was most active in 200 µM and likely for this reason, its effect was less pronounced and not statistically signiﬁcant in the screening.

26 Figure 10: A) Cytokinin derivatives protect C. elegans against oxidative (juglone 500 µM, 5 h) and heat stress (35◦C). B) General structures of majority of active compounds.

What makes these results even more promising is that we have recently shown that some of these derivatives also have a protective effect in models of mitochondrial disease ([1], unpublished). Together with the fact that they show only marginal toxicity to human non-cancer cell lines [103], we believe they might be an interesting candidates for both drug and cosmetics development. Another example of active compounds we discovered during our investigation of cytokinin derivatives are aromatic cytokinins with a 9-(tetrahydrofuran-2-yl) moiety. The preparation and activity of these compounds are described in detail in here [104] (supplement 3). 13 compounds were prepared and tested. We found that two of them, namely 6-(thiophen-2-yl methylamino)-9- (tetrahydrofuran-2-yl)purine and 2-chloro-6-furfurylamino-9-(tetrahydrofuran-2-yl)purine, as well as 6-furfurylamino-9-tetrahydrofuran-2-yl (Kin-THF), were able to signiﬁcantly increase the stress resistance of worms (Figure 11). Their protective effect was not limited to a single model, as these compounds also protected human skin cells against UVA and UVB irradiation, and they were non-toxic in various human skin cell lines. Although their mechanism of action has not yet been investigated, the lack of activity in ORAC assay suggests that the derivatives do not act as direct antioxidants.

27 Figure 11: A) Aromatic cytokinin derivatives with tetrahydrofuran moiety protect C. elegans against oxidative stress. All three compounds signiﬁcantly increased survival (p < 0.05; Log-rank test) after exposure to ROS generator (juglon 500 µM). B) Structures of active derivatives.

28 5 Conclusions and future perspectives

We have established that C. elegans is a suitable model for investigating the protective and pro- longevity effect of cytokinins. We discovered that 3 natural aromatic cytokinin bases – kinetin, ortho-topolin and para-topolin – prolong the lifespan of C. elegans. K, which was selected as a representative compound for follow-up experiments, also increased the thermotolerance and resistance of worms to oxidative stress. It was efficiently absorbed and metabolized by worms into kinetin riboside and kinetin riboside-5’-monophosphate. We also showed that the effect of K was DAF-16 independent, but, interestingly, the presence of ROS seems to be required for its activity. Although these results do not allow us to give a definite answer to the question of how K induces longevity and stress resistance in worms, they permit us to form several plausible hypotheses and provide a basis for follow-up study, in which the involvement of SKN-1 (worm homolog of Nrf2) and AMPK will be investigated. The absence of K in native worm and bacterial samples also prompted us to conduct another follow-up study, in which the presence and metabolism of cytokinins in various species under normal and stress conditions will be examined. We also screened for active synthetic derivatives of natural cytokinin bases using both traditional and modern automated high-throughput screening methods. In our experiments, derivatives with substitution in position 8 of the purine ring and aromatic cytokinins with 9- (tetrahydrofuran-2-yl) moiety seemed to be especially promising. Their protective effect will be further examined in the future. Cytokinins are interesting candidates for medicine and dermatology, and since the 1990s, when their beneficial activity in animal models was described for the first time, dozens of papers regarding their protective effect have been published. However, it has been more than 15 years since the topic was last comprehensively reviewed [137, 138, 139] and these older works understandably do not contain more recent exciting discoveries, such as the effect of K on alternative splicing of pre-mRNA [51] or the ability of KRTP to act as a neo-substrate of disease relevant kinases [49, 50]. This motivated us to summarize current knowledge in a book chapter and a review article on kinetin ([1, 2], supplements 2 and 4).

29 References

[1] A. Kadlecová, B. Maková, M. Artal-Sanz, M. Strnad, and J. Voller, “The plant hormone kinetin in disease therapy and healthy aging,” Ageing Research Reviews, p. 100958, 2019.

[2] J. Voller, B. Maková, A. Kadlecová, G. Gonzalez, and M. Strnad, “Plant hormone cytokinins for modulating human aging and age-related diseases,” in Hormones in Ageing and Longevity, pp. 311–335, Springer, 2017.

[3] C. O. Miller, F. Skoog, M. H. Von Saltza, and F. Strong, “Kinetin, a cell division factor from deoxyribonucleic acid,” Journal of the American Chemical Society, vol. 77, no. 5, pp. 1392–1392, 1955.

[4] D. Letham and C. Miller, “Identity of kinetin-like factors from Zea mays,” Plant and Cell Physiology, vol. 6, no. 2, pp. 355–359, 1965.

[5] N. Raman and S. Elumalai, “Presence of cytokinin in the root nodules of Casuarina equisetifolia,” Indian Journal of Experimental Biology, vol. 34, no. 6, pp. 577–579, 1996.

[6] J. Barciszewski, G. E. Siboska, B. O. Pedersen, B. F. Clark, and S. I. Rattan, “Evidence for the presence of kinetin in DNA and cell extracts,” FEBS letters, vol. 393, no. 2-3, pp. 197–200, 1996.

[7] J. Barciszewski, M. Mielcarek, M. Stobiecki, G. Siboska, and B. F. Clark, “Identiﬁcation of 6- furfuryladenine (kinetin) in human urine,” Biochemical and Biophysical Research Communications, vol. 279, no. 1, pp. 69–73, 2000.

[8] B. Persson, B. Esberg, O. Olafsson, and G. Björk, “Synthesis and function of isopentenyl adenosine derivatives in tRNA,” Biochimie, vol. 76, no. 12, pp. 1152–1160, 1994.

[9] A. Golovko, G. Hjälm, F. Sitbon, and B. Nicander, “Cloning of a human tRNA isopentenyl transferase,” Gene, vol. 258, no. 1-2, pp. 85–93, 2000.

[10] L. Spíchal, “Cytokinins–recent news and views of evolutionally old molecules,” Functional Plant Biology, vol. 39, no. 4, pp. 267–284, 2012.

[11] M. Seegobin, A. Kisiala, A. Noble, D. Kaplan, C. Brunetti, and R. N. Emery, “Canis familiaris tissues are characterized by different proﬁles of cytokinins typical of the tRNA degradation pathway,” The FASEB Journal, vol. 32, no. 12, pp. 6575–6581, 2018.

[12] T. Kakimoto, “Identiﬁcation of plant cytokinin biosynthetic enzymes as dimethylallyl diphosphate: ATP/ADP isopentenyltransferases,” Plant and Cell Physiology, vol. 42, no. 7, pp. 677–685, 2001.

[13] J. J. Kieber and G. E. Schaller, “Cytokinins,” The Arabidopsis Book/American Society of Plant Biologists, vol. 12, 2014.

30 [14] T. Kurakawa, N. Ueda, M. Maekawa, K. Kobayashi, M. Kojima, Y. Nagato, H. Sakakibara, and J. Kyozuka, “Direct control of shoot meristem activity by a cytokinin-activating enzyme,” Nature, vol. 445, no. 7128, p. 652, 2007.

[15] T. Schmülling, T. Werner, M. Rieﬂer, E. Krupková, and I. B. y Manns, “Structure and function of cytokinin oxidase/dehydrogenase genes of maize, rice, Arabidopsis and other species,” Journal of Plant Research, vol. 116, no. 3, pp. 241–252, 2003.

[16] T. Kudo, T. Kiba, and H. Sakakibara, “Metabolism and long-distance translocation of cytokinins,” Journal of Integrative Plant Biology, vol. 52, no. 1, pp. 53–60, 2010.

[17] Y. Durán-Medina, D. Díaz-Ramírez, and N. Marsch-Martínez, “Cytokinins on the move,” Frontiers in Plant Science, vol. 8, p. 146, 2017.

[18] I. Hwang, J. Sheen, and B. Müller, “Cytokinin signaling networks,” Annual Review of Plant Biology, vol. 63, pp. 353–380, 2012.

[19] J. J. Kieber and G. E. Schaller, “Cytokinin signaling in plant development,” Development, vol. 145, no. 4, 2018.

[20] E. A. Keshishian and A. M. Rashotte, “Plant cytokinin signalling,” Essays in biochemistry, vol. 58, pp. 13– 27, 2015.

[21] G. E. Schaller, A. Bishopp, and J. J. Kieber, “The yin-yang of hormones: cytokinin and auxin interactions in plant development,” The Plant Cell, vol. 27, no. 1, pp. 44–63, 2015.

[22] T. Werner and T. Schmülling, “Cytokinin action in plant development,” Current Opinion in Plant Biology, vol. 12, no. 5, pp. 527–538, 2009.

[23] J. Voller, T. Béres, M. Zatloukal, P. Džubák, M. Hajdúch, K. Doležal, T. Schmülling, and S. Miroslav, “Anti-cancer activities of cytokinin ribosides,” Phytochemistry Reviews, pp. 1–13, 2019.

[24] S. I. Rattan and B. F. Clark, “Kinetin delays the onset of aging characteristics in human ﬁbroblasts,” Bio- chemical and biophysical research communications, vol. 201, no. 2, pp. 665–672, 1994.

[25] S. P. Sharma, P. Kaur, and S. I. Rattan, “Plant growth hormone kinetin delays aging, prolongs the lifespan, and slows down development of the fruitﬂy Zaprionus paravittiger,” Biochemical and Biophysical Research Communications, vol. 216, no. 3, pp. 1067–1071, 1995.

[26] S. P. Sharma, J. Kaur, and S. I. Rattan, “Increased longevity of kinetin-fed Zaprionus fruitﬂies is accompanied by their reduced fecundity and enhanced catalase activity,” IUBMB Life, vol. 41, no. 5, pp. 869–875, 1997.

[27] A. Kadlecová, T. Jirsa, O. Novák, J. Kammenga, M. Strnad, and J. Voller, “Natural plant hormones cytokinins increase stress resistance and longevity of Caenorhabditis elegans,” Biogerontology, vol. 19, no. 2, pp. 109–120, 2018.

31 [28] T. Kimura and K. Doi, “Depigmentation and rejuvenation effects of kinetin on the aged skin of hairless descendants of Mexican hairless dogs,” Rejuvenation Research, vol. 7, no. 1, pp. 32–39, 2004.

[29] J. McCullough and G. Weinstein, “Clinical study of safety and efﬁcacy of using topical kinetin 0.1%

(kinerase R ) to treat photodamaged skin,” Cosmetic Dermatology, vol. 15, pp. 29–32, 09 2002.

[30] R. Wanitphakdeedecha, W. Meeprathom, W. Manuskiatti, et al., “Efﬁcacy and safety of 0.1% kinetin cream in the treatment of photoaging skin,” Indian Journal of Dermatology, Venereology, and Leprology, vol. 81, no. 5, p. 547, 2015.

[31] P.-C. Chiu, C.-C. Chan, H.-M. Lin, and H.-C. Chiu, “The clinical anti-aging effects of topical kinetin and niacinamide in asians: a randomized, double-blind, placebo-controlled, split-face comparative trial,” Jour- nal of Cosmetic Dermatology, vol. 6, no. 4, pp. 243–249, 2007.

[32] D. McDaniel, B. Neudecker, J. DiNardo, J. Lewis, and H. Maibach, “Idebenone: a new antioxidant–part i. relative assessment of oxidative stress protection capacity compared to commonly known antioxidants,” Journal of Cosmetic Dermatology, vol. 4, no. 1, pp. 10–17, 2005.

[33] S. An, H. J. Cha, J.-M. Ko, H. Han, S. Y. Kim, K.-S. Kim, S. J. Lee, I.-S. An, S. Kim, H. J. Youn, et al., “Kinetin improves barrier function of the skin by modulating keratinocyte differentiation markers,” Annals of Dermatology, vol. 29, no. 1, pp. 6–12, 2017.

[34] S. I. Rattan and L. Sodagam, “Gerontomodulatory and youth-preserving effects of zeatin on human skin ﬁbroblasts undergoing aging in vitro,” Rejuvenation Research, vol. 8, no. 1, pp. 46–57, 2005.

[35] C. Ji, Y. Yang, B. Yang, J. Xia, W. Sun, Z. Su, L. Yu, S. Shan, S. He, L. Cheng, et al.,“Trans-zeatin attenuates ultraviolet induced down-regulation of aquaporin-3 in cultured human skin keratinocytes,” Inter- national Journal of Molecular Medicine, vol. 26, no. 2, pp. 257–263, 2010.

[36] B. Yang, C. Ji, J. Kang, W. Chen, Z. Bi, and Y. Wan, “Trans-zeatin inhibits UVB-induced matrix metalloproteinase-1 expression via map kinase signaling in human skin ﬁbroblasts,” International Journal of Molecular Medicine, vol. 23, no. 4, pp. 555–560, 2009.

[37] J. L. McCullough, R. L. Garcia, and B. Reece, “A clinical study of topical Pyratine 6 for improving the appearance of photodamaged skin.,” Journal of Drugs in Dermatology: JDD, vol. 7, no. 2, pp. 131–135, 2008.

[38] A. M. Tremaine, A. Ortiz, L. Elkeeb, M. Tran, and G. Weinstein, “Long-term efﬁcacy and safety of topical PRK 124 (0.125%) lotion (Pyratine-XR) in the treatment of mild-to-moderate rosacea.,” Journal of Drugs in Dermatologyy: JDD, vol. 9, no. 6, pp. 647–650, 2010.

[39] A. Ortiz, L. Elkeeb, A. Truitt, R. Hindiyeh, L. Aquino, M. Tran, and G. Weinstein, “Topical PRK 124 (0.125%) lotion for improving the signs and symptoms of rosacea.,” Journal of drugs in dermatology: JDD, vol. 8, no. 5, pp. 459–462, 2009.

32 [40] U. Berge, P. Kristensen, and S. I. Rattan, “Kinetin-induced differentiation of normal human keratinocytes undergoing aging in vitro,” Annals of the New York Academy of Sciences, vol. 1067, no. 1, pp. 332–336, 2006.

[41] U. Berge, P. Kristensen, and S. I. Rattan, “Hormetic modulation of differentiation of normal human epidermal keratinocytes undergoing replicative senescence in vitro,” Experimental Gerontology, vol. 43, no. 7, pp. 658–662, 2008.

[42] L. Bolund, P. K. Jensen, and P. Bjerring, “Method and composition for treating hyperproliferative skin diseases using 6-aminopurine cytokinins,” June 4 1991. US Patent 5,021,422.

[43] Y. Wei, D. Liu, Y. Zheng, C. Hao, H. Li, and W. Ouyang, “Neuroprotective effects of kinetin against glutamate-induced oxidative cytotoxicity in HT22 cells: Involvement of Nrf2 and Heme Oxygenase-1,” Neurotoxicity Research, vol. 33, no. 4, pp. 725–737, 2018.

[44] Y. Liu, Z. Zhang, and X. Yang, “Kinetin protects against lipid peroxidation and improves antioxidant status in cultured astrocytes and mouse brain exposed to D-galactose,” African Journal of Biotechnology, vol. 10, no. 55, pp. 11721–11727, 2011.

[45] V. Radhakrishna, S. K. Nanilu, G. Sanjeev, J. Shetty, Y. T. Somyaji, and S. S. Moodithaya, “Evaluation of the potency of kinetin on radiation induced behavioural changes in Swiss Albino Mice,” Journal of Clinical and Diagnostic Research: JCDR, vol. 11, no. 7, p. TF01, 2017.

[46] Y. Wei, D. Liu, Y. Zheng, H. Li, C. Hao, and W. Ouyang, “Protective effects of kinetin against aluminum chloride and D-galactose induced cognitive impairment and oxidative damage in mouse,” Brain Research Bulletin, vol. 134, pp. 262–272, 2017.

[47] S. J. Choi, C.-H. Jeong, S.-G. Choi, J.-Y. Chun, Y. J. Kim, J. Lee, D.-H. Shin, and H. J. Heo, “Zeatin prevents amyloid β-induced neurotoxicity and scopolamine-induced cognitive deﬁcits,” Journal of Medicinal Food, vol. 12, no. 2, pp. 271–277, 2009.

[48] M. J. Kim, S. J. Choi, S.-T. Lim, H. K. Kim, Y. J. Kim, H.-G. Yoon, and D.-H. Shin, “Zeatin supplement improves scopolamine-induced memory impairment in mice,” Bioscience, Biotechnology, and Biochemistry, vol. 72, no. 2, pp. 577–581, 2008.

[49] N. T. Hertz, A. Berthet, M. L. Sos, K. S. Thorn, A. L. Burlingame, K. Nakamura, and K. M. Shokat, “A neo-substrate that ampliﬁes catalytic activity of Parkinson’s-disease-related kinase PINK1,” Cell, vol. 154, no. 4, pp. 737–747, 2013.

[50] L. E. Bowie, T. Maiuri, M. Alpaugh, M. Gabriel, N. Arbez, D. Galleguillos, C. L. Hung, S. Patel, J. Xia, N. T. Hertz, et al., “N6-furfuryladenine is protective in Huntington’s disease models by signaling huntingtin phosphorylation,” Proceedings of the National Academy of Sciences, vol. 115, no. 30, pp. E7081–E7090, 2018.

33 [51] S. A. Slaugenhaupt, J. Mull, M. Leyne, M. P. Cuajungco, S. P. Gill, M. M. Hims, F. Quintero, F. B. Axel- rod, and J. F. Gusella, “Rescue of a human mRNA splicing defect by the plant cytokinin kinetin,” Human Molecular Genetics, vol. 13, no. 4, pp. 429–436, 2004.

[52] M. Shohat and M. W. Hubshman, “Familial dysautonomia,” GeneReviews, PMID: 20301359, 2014.

[53] M. M. Hims, M. Leyne, J. Mull, L. Liu, C. Lazaro, R. S. Shetty, S. Gill, J. F. Gusella, R. Reed, S. A. Slaugenhaupt, et al., “Therapeutic potential and mechanism of kinetin as a treatment for the human splicing disease familial dysautonomia,” Journal of Molecular Medicine, vol. 85, no. 2, pp. 149–161, 2007.

[54] R. Sinha, Y. J. Kim, T. Nomakuchi, K. Sahashi, Y. Hua, F. Rigo, C. F. Bennett, and A. R. Krainer, “Antisense oligonucleotides correct the familial dysautonomia splicing defect in IKBKAP transgenic mice,” Nucleic Acids Research, vol. 46, no. 10, pp. 4833–4844, 2018.

[55] N. Zeltner, F. Fattahi, N. C. Dubois, N. Saurat, F. Lafaille, L. Shang, B. Zimmer, J. Tchieu, M. A. Soliman, G. Lee, et al., “Capturing the biology of disease severity in a PSC-based model of familial dysautonomia,” Nature Medicine, vol. 22, no. 12, p. 1421, 2016.

[56] G. Lee, E. P. Papapetrou, H. Kim, S. M. Chambers, M. J. Tomishima, C. A. Fasano, Y. M. Ganat, J. Menon, F. Shimizu, A. Viale, et al., “Modelling pathogenesis and treatment of familial dysautonomia using patient- speciﬁc iPSCs,” Nature, vol. 461, no. 7262, p. 402, 2009.

[57] N. Boone, B. Loriod, A. Bergon, O. Sbai, C. Formisano-Tréziny, J. Gabert, M. Khrestchatisky, C. Nguyen, F. Féron, F. B. Axelrod, et al., “Olfactory stem cells, a new cellular model for studying molecular mechanisms underlying familial dysautonomia,” PLoS One, vol. 5, no. 12, p. e15590, 2010.

[58] R. S. Shetty, C. S. Gallagher, Y.-T. Chen, M. M. Hims, J. Mull, M. Leyne, J. Pickel, D. Kwok, and S. A. Slau- genhaupt, “Speciﬁc correction of a splice defect in brain by nutritional supplementation,” Human Molecular Genetics, vol. 20, no. 21, pp. 4093–4101, 2011.

[59] G. Gold-von Simson, J. D. Goldberg, L. M. Rolnitzky, J. Mull, M. Leyne, A. Voustianiouk, S. A. Slaugen- haupt, and F. B. Axelrod, “Kinetin in familial dysautonomia carriers: implications for a new therapeutic strategy targeting mRNA splicing,” Pediatric Research, vol. 65, no. 3, p. 341, 2009.

[60] N. Bakhiya and K. E. Appel, “Toxicity and carcinogenicity of furan in human diet,” Archives of Toxicology, vol. 84, no. 7, pp. 563–578, 2010.

[61] E. Pros, J. Fernández-Rodríguez, L. Benito, A. Ravella, G. Capellá, I. Blanco, E. Serra, and C. Lázaro, “Modulation of aberrant NF1 pre-mRNA splicing by kinetin treatment,” European Journal of Human Ge- netics, vol. 18, no. 5, p. 614, 2010.

[62] M. Li, W. Ouyang, X. Wu, Y. Zheng, Y. Wei, and L. An, “Kinetin inhibits apoptosis of aging spleen cells induced by D-galactose in rats,” Journal of Veterinary Science, vol. 15, no. 3, pp. 353–359, 2014.

34 [63] M. Li, W. Ouyang, X. Wu, Y. Zheng, R. Gao, and J. Tang, “Effect of kinetin on immunity and splenic lymphocyte proliferation in vitro in D-galactose-induced aging rats,” Sheng li xue bao:[Acta Physiologica Sinica], vol. 66, no. 5, pp. 605–611, 2014.

[64] M. Li, W. Ouyang, X. Wu, Y. Zheng, R. Gao, X. Ouyang, L. An, et al., “Kinetin protects D-galactose- induced aging rats from splenic oxidative damage,” Pak Vet J, vol. 34, pp. 386–390, 2014.

[65] J. Sun, Y. Liu, T. Cao, and W. Ouyang, “Effect of kinetin on ovary and uterus in D-galactose-induced female mouse model of aging,” Sheng li xue bao:[Acta Physiologica Sinica], vol. 65, no. 4, pp. 389–394, 2013.

[66] J. H. Lee, K. Y. Chung, D. Bang, and K. H. Lee, “Searching for aging-related proteins in human dermal microvascular endothelial cells treated with anti-aging agents,” Proteomics, vol. 6, no. 4, pp. 1351–1361, 2006.

[67] G. Hsiao, M.-Y. Shen, K.-H. Lin, C.-Y. Chou, N.-H. Tzu, C.-H. Lin, D.-S. Chou, T.-F. Chen, and J.-R. Sheu, “Inhibitory activity of kinetin on free radical formation of activated platelets in vitro and on thrombus formation in vivo,” European Journal of Pharmacology, vol. 465, no. 3, pp. 281–287, 2003.

[68] J. R. Sheu, G. Hsiao, M. Shen, C. Chou, C. Lin, T.-F. Chen, and D. Chou, “Inhibitory mechanisms of kinetin, a plant growth-promoting hormone, in platelet aggregation,” Platelets, vol. 14, no. 3, pp. 189–196, 2003.

[69] D. Mijikovic, J. Hranisavljevic, and Z. Pietrzkowski, “Pharmaceutical compositions and methods for metabolic modulation,” July 12 2007. US Patent App. 10/567,875.

[70] A. Olsen, G. E. Siboska, B. F. Clark, and S. I. Rattan, “N6-furfuryladenine, kinetin, protects against Fenton reaction-mediated oxidative damage to DNA,” Biochemical and Biophysical Research Communications, vol. 265, no. 2, pp. 499–502, 1999.

[71] P. Verbeke, G. E. Siboska, B. F. Clark, and S. I. Rattan, “Kinetin inhibits protein oxidation and glycoxidation in vitro,” Biochemical and Biophysical Research Communications, vol. 276, no. 3, pp. 1265–1270, 2000.

[72] A. Brizzolari, C. Marinello, M. Carini, E. Santaniello, and P. A. Biondi, “Evaluation of the antioxidant activity and capacity of some natural N6-substituted adenine derivatives (cytokinins) by ﬂuorimetric and spectrophotometric assays,” Journal of Chromatography B, vol. 1019, pp. 164–168, 2016.

[73] E. M. Othman, M. Naseem, E. Awad, T. Dandekar, and H. Stopper, “The plant hormone cytokinin confers protection against oxidative stress in mammalian cells,” PloS one, vol. 11, no. 12, p. e0168386, 2016.

[74] S. I. Rattan, “N6-furfuryladenine (kinetin) as a potential anti-aging molecule,” Journal of Anti-aging Medicine, vol. 5, no. 1, pp. 113–116, 2002.

[75] S. I. Rattan, “Principles and practice of hormetic treatment of aging and age-related diseases,” Human & Experimental Toxicology, vol. 27, no. 2, pp. 151–154, 2008.

35 [76] M. Mattson, “Hormesis deﬁned,” Ageing Research Reviews, vol. 7, no. 1, pp. 1–7, 2008.

[77] A. Barrière and M.-A. Félix, “Isolation of C. elegans and related nematodes,” WormBook, vol. 17, pp. 1–9, 2006.

[78] S. Brenner, “The genetics of Caenorhabditis elegans,” Genetics, vol. 77, no. 1, pp. 71–94, 1974.

[79] J. E. Sulston and H. R. Horvitz, “Post-embryonic cell lineages of the nematode, Caenorhabditis elegans,” Developmental Biology, vol. 56, no. 1, pp. 110–156, 1977.

[80] J. E. Sulston, E. Schierenberg, J. G. White, and J. Thomson, “The embryonic cell lineage of the nematode Caenorhabditis elegans,” Developmental Biology, vol. 100, no. 1, pp. 64–119, 1983.

[81] A. Fire, S. Xu, M. K. Montgomery, S. A. Kostas, S. E. Driver, and C. C. Mello, “Potent and speciﬁc genetic interference by double-stranded RNA in Caenorhabditis elegans,” Nature, vol. 391, no. 6669, p. 806, 1998.

[82] M. Chalﬁe, Y. Tu, G. Euskirchen, W. W. Ward, and D. C. Prasher, “Green ﬂuorescent protein as a marker for gene expression,” Science, vol. 263, no. 5148, pp. 802–805, 1994.

[83] J. Spieth, D. Lawson, P. Davis, G. Williams, and K. Howe, “Overview of gene structure in C. elegans,” WormBook, 2014.

[84] T. C. elegans Sequencing Consortium, “Genome sequence of the nematode C. elegans: a platform for investigating biology,” Science, pp. 2012–2018, 1998.

[85] E. Culetto and D. B. Sattelle, “A role for Caenorhabditis elegans in understanding the function and interactions of human disease genes,” Human Molecular Genetics, vol. 9, no. 6, pp. 869–877, 2000.

[86] T. Kaletta and M. O. Hengartner, “Finding function in novel targets: C. elegans as a model organism,” Nature Reviews Drug Discovery, vol. 5, no. 5, p. 387, 2006.

[87] L. P. O’Reilly, C. J. Luke, D. H. Perlmutter, G. A. Silverman, and S. C. Pak, “C. elegans in high-throughput drug discovery,” Advanced Drug Delivery Reviews, vol. 69, pp. 247–253, 2014.

[88] C. López-Otín, M. A. Blasco, L. Partridge, M. Serrano, and G. Kroemer, “The hallmarks of aging,” Cell, vol. 153, no. 6, pp. 1194–1217, 2013.

[89] J. Aunan, M. Watson, H. Hagland, and K. Søreide, “Molecular and biological hallmarks of ageing,” British Journal of Surgery, vol. 103, no. 2, pp. e29–e46, 2016.

[90] M. Klass and D. Hirsh, “Non-ageing developmental variant of Caenorhabditis elegans,” Nature, vol. 260, no. 5551, p. 523, 1976.

[91] H. A. Tissenbaum, “Using C. elegans for aging research,” Invertebrate Reproduction & Development, vol. 59, no. sup1, pp. 59–63, 2015.

36 [92] J. J. Collins, C. Huang, S. Hughes, and K. Kornfeld, “The measurement and analysis of age-related changes in Caenorhabditis elegans,” WormBook, 2007.

[93] D. B. Friedman and T. E. Johnson, “A mutation in the age-1 gene in Caenorhabditis elegans lengthens life and reduces hermaphrodite fertility.,” Genetics, vol. 118, no. 1, pp. 75–86, 1988.

[94] C. Kenyon, J. Chang, E. Gensch, A. Rudner, and R. Tabtiang, “A C. elegans mutant that lives twice as long as wild type,” Nature, vol. 366, no. 6454, p. 461, 1993.

[95] M. Barbieri, M. Bonafè, C. Franceschi, and G. Paolisso, “Insulin/IGF-I-signaling pathway: an evolutionarily conserved mechanism of longevity from yeast to humans,” American Journal of Physiology- Endocrinology And Metabolism, vol. 285, no. 5, pp. E1064–E1071, 2003.

[96] M. Uno and E. Nishida, “Lifespan-regulating genes in C. elegans,” NPJ Aging and Mechanisms of Disease, vol. 2, p. 16010, 2016.

[97] M. W ˛atrobaand D. Szukiewicz, “The role of sirtuins in aging and age-related diseases,” Advances in Med- ical Sciences, vol. 61, no. 1, pp. 52–62, 2016.

[98] V. D. Longo and B. K. Kennedy, “Sirtuins in aging and age-related disease,” Cell, vol. 126, no. 2, pp. 257– 268, 2006.

[99] R. A. Saxton and D. M. Sabatini, “mTOR signaling in growth, metabolism, and disease,” Cell, vol. 168, no. 6, pp. 960–976, 2017.

[100] S.-M. Jeon, “Regulation and function of AMPK in physiology and diseases,” Experimental & Molecular Medicine, vol. 48, no. 7, p. e245, 2016.

[101] H. Pan and T. Finkel, “Key proteins and pathways that regulate lifespan,” Journal of Biological Chemistry, vol. 292, no. 16, pp. 6452–6460, 2017.

[102] C. J. Kenyon, “The genetics of ageing,” Nature, vol. 464, no. 7288, p. 504, 2010.

[103] L. Zahajská, J. Nisler, J. Voller, T. Gucký, T. Pospíšil, L. Spíchal, and M. Strnad, “Preparation, characterization and biological activity of C8-substituted cytokinins,” Phytochemistry, vol. 135, pp. 115–127, 2017.

[104] M. Hönig, L. Plíhalová, L. Spíchal, J. Grúz, A. Kadlecová, J. Voller, A. R. Svobodová, J. Vostálová, J. Ul- richová, K. Doležal, et al., “New cytokinin derivatives possess UVA and UVB photoprotective effect on human skin cells and prevent oxidative stress,” European Journal of Medicinal Chemistry, vol. 150, pp. 946– 957, 2018.

[105] T. Stiernagle, “Maintenance of C. elegans,” C. elegans, vol. 2, pp. 51–67, 1999.

[106] B. R. Ellerbrock, E. M. Coscarelli, M. E. Gurney, and T. G. Geary, “Screening for presenilin inhibitors using the free-living nematode, Caenorhabditis elegans,” Journal of Biomolecular Screening, vol. 9, no. 2, pp. 147–152, 2004.

37 [107] G. Solis and M. Petrascheck, “Measuring Caenorhabditis elegans life span in 96 well microtiter plates,” Journal of Visualized Experiments: JoVE, no. 49, 2011.

[108] B. Hernando-Rodríguez, A. P. Erinjeri, M. J. Rodríguez-Palero, V. Millar, S. González-Hernández, M. Olmedo, B. Schulze, R. Baumeister, M. J. Muñoz, P. Askjaer, et al., “Combined ﬂow cytometry and high-throughput image analysis for the study of essential genes in Caenorhabditis elegans,” BMC Biology, vol. 16, no. 1, p. 36, 2018.

[109] A. Brizzolari, C. Marinello, M. Carini, E. Santaniello, and P. Biondi, “Evaluation of the antioxidant activity and capacity of some natural N6-substituted adenine derivatives (cytokinins) by ﬂuorimetric and spectrophotometric assays,” Journal of Chromatography B, vol. 1019, pp. 164–168, 2016.

[110] R. Garcia, J. McCullough, B. Reece, and F. Massino, “Novel topical cytokinin improves coarse wrinkles and skin roughness without skin irritation.,” Journal of Drugs in Dermatology: JDD, vol. 17, no. 7, pp. 772–779, 2018.

[111] K. Zhou, Z. Pincus, and F. Slack, “Longevity and stress in Caenorhabditis elegans,” Aging (Albany NY), vol. 3, pp. 733–753, Aug 2011.

[112] J. J. Inbaraj and C. F. Chignell, “Cytotoxic action of juglone and plumbagin: a mechanistic study using HaCaT keratinocytes,” Chemical Research in Toxicology, vol. 17, no. 1, pp. 55–62, 2004.

[113] C. Kenyon, J. Chang, E. Gensch, A. Rudner, R. Tabtiang, et al., “A C. elegans mutant that lives twice as long as wild type,” Nature, vol. 366, no. 6454, pp. 461–464, 1993.

[114] D. Harman, “Aging: a theory based on free radical and radiation chemistry,” Journal of Gerontology, vol. 11, no. 3, pp. 298–300, 1955.

[115] P. Davalli, T. Mitic, A. Caporali, A. Lauriola, and D. D’Arca, “ROS, cell senescence, and novel molecular mechanisms in aging and age-related diseases,” Oxidative Medicine and Cellular Longevity, vol. 2016, pp. 1–18, 2016.

[116] M. Ristow and K. Schmeisser, “Mitohormesis: promoting health and lifespan by increased levels of reactive oxygen species (ROS),” Dose-Response, vol. 12, no. 2, pp. dose–response, 2014.

[117] J. Barciszewski, G. E. Siboska, B. O. Pedersen, B. F. Clark, and S. I. Rattan, “A mechanism for the in vivo formation of N6-furfuryladenine, kinetin, as a secondary oxidative damage product of DNA,” FEBS letters, vol. 414, no. 2, pp. 457–460, 1997.

[118] J. Barciszewski, G. E. Siboska, B. O. Pedersen, B. F. Clark, and S. I. Rattan, “Furfural, a precursor of the cytokinin hormone kinetin, and base propenals are formed by hydroxyl radical damage of DNA,” Biochemical and Biophysical Research Communications, vol. 238, no. 2, pp. 317–319, 1997.

38 [119] E. Wyszko, M. Z. Barciszewska, M. Markiewicz, M. Szymanski,´ W. T. Markiewicz, B. F. Clark, and J. Bar- ciszewski, ““Action-at-a distance” of a new DNA oxidative damage product 6-furfuryl-adenine (kinetin) on template properties of modiﬁed DNA,” Biochimica et Biophysica Acta (BBA)-Gene Structure and Expres- sion, vol. 1625, no. 3, pp. 239–245, 2003.

[120] L. Ge, J. W. H. Yong, N. K. Goh, L. S. Chia, S. N. Tan, and E. S. Ong, “Identiﬁcation of kinetin and kinetin riboside in coconut (Cocos nucifera L.) water using a combined approach of liquid chromatography– tandem mass spectrometry, high performance liquid chromatography and capillary electrophoresis,” Journal of Chromatography B, vol. 829, no. 1-2, pp. 26–34, 2005.

[121] A. L. Orr, F. U. Rutaganira, D. de Roulet, E. J. Huang, N. T. Hertz, K. M. Shokat, and K. Nakamura, “Long-term oral kinetin does not protect against α-synuclein-induced neurodegeneration in rodent models of Parkinson’s disease,” Neurochemistry International, vol. 109, pp. 106–116, 2017.

[122] Y. Ishii, Y. Hori, S. Sakai, and Y. Honma, “Control of differentiation and apoptosis of human myeloid leukemia cells by cytokinins and cytokinin nucleosides, plant redifferentiation-inducing hormones,” Cell Growth and Differentiation, vol. 13, no. 1, pp. 19–26, 2002.

[123] C. Cabello, W. Bair, S. Ley, S. Lamore, S. Azimian, and G. Wondrak, “The experimental chemotherapeutic N6-furfuryladenosine (kinetin-riboside) induces rapid ATP depletion, genotoxic stress, and CDKN1A (p21) upregulation in human cancer cell lines,” Biochemical Pharmacology, vol. 77, no. 7, pp. 1125–1138, 2009.

[124] J. Cheong, D. Goh, J. Yong, S. Tan, and E. Ong, “Inhibitory effect of kinetin riboside in human heptamoa, HepG2,” Molecular BioSystems, vol. 5, no. 1, pp. 91–98, 2009.

[125] A. Dassano, M. Mancuso, P. Giardullo, L. De Cecco, P. Ciuffreda, E. Santaniello, A. Saran, T. A. Dragani, and F. Colombo, “N6-isopentenyladenosine and analogs activate the NRF2-mediated antioxidant response,” Redox Biology, vol. 2, pp. 580–589, 2014.

[126] Q. Ma, “Role of Nrf2 in oxidative stress and toxicity,” Annual Review of Pharmacology and Toxicology, vol. 53, pp. 401–426, 2013.

[127] L. A. Peterson, “Reactive metabolites in the biotransformation of molecules containing a furan ring,” Chem- ical Research in Toxicology, vol. 26, no. 1, pp. 6–25, 2012.

[128] D. Mijikovic, J. Hranisavljevic, and Z. Pietrzkowski, “Pharmaceutical compositions and methods for metabolic modulation,” Aug. 5 2004. US Patent App. 10/567,875.

[129] S. Pisanti, P. Picardi, E. Ciaglia, L. Margarucci, R. Ronca, A. Giacomini, A. M. Malﬁtano, A. Casapullo, C. Laezza, P. Gazzerro, et al., “Antiangiogenic effects of N6-isopentenyladenosine, an endogenous isoprenoid end product, mediated by AMPK activation,” The FASEB Journal, vol. 28, no. 3, pp. 1132–1144, 2014.

39 [130] P. Mlejnek and P. Doležel, “Apoptosis induced by N6-substituted derivatives of adenosine is related to intracellular accumulation of corresponding mononucleotides in HL-60 cells,” Toxicology in vitro, vol. 19, no. 7, pp. 985–990, 2005.

[131] T. Schulz, K. Zarse, A. Voigt, N. Urban, M. Birringer, and M. Ristow, “Glucose restriction extends Caenorhabditis elegans life span by inducing mitochondrial respiration and increasing oxidative stress,” Cell Metabolism, vol. 6, no. 4, pp. 280–293, 2007.

[132] P. Mungai, G. Waypa, A. Jairaman, M. Prakriya, D. Dokic, M. Ball, and P. Schumacker, “Hypoxia triggers AMPK activation through reactive oxygen species-mediated activation of calcium release-activated calcium channels,” Molecular and Cellular Biology, vol. 31, no. 17, pp. 3531–3545, 2011.

[133] W. De Haes, L. Frooninckx, R. Van Assche, A. Smolders, G. Depuydt, J. Billen, B. Braeckman, L. Schoofs, and L. Temmerman, “Metformin promotes lifespan through mitohormesis via the peroxiredoxin PRDX-2,” Proceedings of the National Academy of Sciences, vol. 111, no. 24, pp. E2501–E2509, 2014.

[134] J. Apfeld, G. O’Connor, T. McDonagh, P. S. DiStefano, and R. Curtis, “The AMP-activated protein kinase AAK-2 links energy levels and insulin-like signals to lifespan in C. elegans,” Genes & Development, vol. 18, no. 24, pp. 3004–3009, 2004.

[135] M. G. Novelle, A. Ali, C. Diéguez, M. Bernier, and R. de Cabo, “Metformin: a hopeful promise in aging research,” Cold Spring Harbor Perspectives in Medicine, vol. 6, no. 3, p. a025932, 2016.

[136] B. Onken and M. Driscoll, “Metformin induces a dietary restriction–like state and the oxidative stress response to extend C. elegans healthspan via AMPK, LKB1, and SKN-1,” PloS One, vol. 5, no. 1, p. e8758, 2010.

[137] S. Rattan, “N6-furfuryladenine (kinetin) as a potential anti-aging molecule,” Journal of Anti-aging Medicine, vol. 5, no. 1, pp. 113–116, 2002.

[138] J. Barciszewski, S. I. Rattan, G. Siboska, and B. F. Clark, “Kinetin—45 years on,” Plant Science, vol. 148, no. 1, pp. 37–45, 1999.

[139] J. Barciszewski, G. Siboska, S. I. Rattan, and B. F. Clark, “Occurrence, biosynthesis and properties of kinetin (N6-furfuryladenine),” Plant Growth Regulation, vol. 32, no. 2-3, pp. 257–265, 2000.

40 6 List of published papers and other contributions

Papers in journals with impact factor

1. Kadlecová, A., Jirsa, T., Novák, O., Kammenga, J., Strnad, M., & Voller, J. (2018). Nat- ural plant hormones cytokinins increase stress resistance and longevity of Caenorhabditis elegans. Biogerontology, 19(2), 109-120.

2. Kadlecová, A., Maková, B., Artal-Sanz, M., Strnad, M. & Voller, J. (2019). The plant hormone kinetin in disease therapy and healthy aging. Ageing Research Reviews, in press.

3. Hönig, M., Plíhalová, L., Spíchal, L., Grúz, J., Kadlecová, A., Voller, J., Svobodová, A.R., Vostálová, J., Ulrichová, J., Doležal, K., & Strnad, M. (2018). New cytokinin derivatives possess UVA and UVB photoprotective effect on human skin cells and prevent oxidative stress. European journal of medicinal chemistry, 150, 946-957.

Book chapter

1. Voller, J., Maková, B., Kadlecová, A., Gonzalez, G., & Strnad, M. (2017). Plant hormone cytokinins for modulating human aging and age-related diseases. In Hormones in Ageing and Longevity (pp. 311-335). Springer, Cham.

Conference contributions

1. Kadlecová, A., Jirsa, T., Hönig, M., Novák, O., Voller, J., Plíhalová, L., Doležal, K., Strnad, M. Protective effect of plant hormones cytokinins in Caenorhabditis elegans. EMBO workshop: C. elegans development, cell biology and gene expression, Barcelona 2018. Poster.

2. Kadlecová, A. Protective effect of cytokinins and their derivatives in Caenorhabditis elegans / News from the „worm cabinet“. CBPRS 2018, Luhacoviceˇ 2018. Oral presentation.

41 3. Kadlecová, A., Jirsa, T., Strnad, M., Voller, J. Cytoprotective activity of natural cytokinin bases in Caenorhabditis elegans. Trends in Natural Product Research – PSE Young Scientists’ Meeting, Natural Products in Health, Agro-Food and Cosmetics, Lille 2017. Poster.

4. Voller, J., Kadlecová, A., Maková, B., Zahájská, L., Plíhalová, L., Grúz, J., Zatloukal, M., Schubert, D., Spíchal, L., Strnad, M. Antiaging activity of plant hormones cytokinins. The future of aging, Warsaw 2016. Poster.

5. Kadlecová, A. Worms getting high-throughput. Growth regulators on the way, Malá Morávka 2016. Oral presentation.

6. Voller, J., Kadlecová, A.. Databáze látek zpomalujících stárnutí. Národní bioinformat- ická konference ENBIK, Louceˇ n,ˇ 2016. Poster.

7. Kadlecová, A., Schubert, D., Voller, J., Strnad, M. Caenorhabditis elegans as a model organism for testing of anti-aging activity of plant hormones cytokinins. Systems Biology of Ageing, Jena 2015. Poster.

8. Voller, J., Plíhalová, L., Zahájská, L., Zatloukal, M., Kadlecová, A., Grúz, J., Nardel- liová, B., Schubert, D., Spíchal, L., Strnad, M. Anti-aging activity of phytohormones cytokinins. Healthy Aging: From Molecules to Organisms, Hinxton 2015. Poster.

9. Kadlecová, A. Studying cytoprotective activity of cytokinins in Caenorhabditis elegans. Biotechnology of Phytohormones and Natural Substances, Malá Morávka 2015. Oral presentation.

Awarded grants

1. INTER-COST no. LTC17071 – Hodnocení úcink˚urostlinnýchˇ hormon˚ua jejich derivát˚u na hád’átko obecné. Principal investigator: Mgr. Alena Kadlecová. Duration: 2017 – 2019.

42 7 Supplements

7.1 Supplement 1

Kadlecová, A., Jirsa, T., Novák, O., Kammenga, J., Strnad, M., & Voller, J. (2018). Natural plant hormones cytokinins increase stress resistance and longevity of Caenorhabditis elegans. Biogerontology, 19(2), 109-120.

43 Biogerontology (2018) 19:109–120 https://doi.org/10.1007/s10522-017-9742-4

RESEARCH ARTICLE

Natural plant hormones cytokinins increase stress resistance and longevity of Caenorhabditis elegans

Alena Kadlecová . Tomáš Jirsa . Ondřej Novák . Jan Kammenga . Miroslav Strnad . Jiří Voller

Received: 20 October 2017 / Accepted: 14 December 2017 / Published online: 18 December 2017 © Springer Science+Business Media B.V., part of Springer Nature 2017

Abstract Cytokinins are phytohormones that are in cosmetics. To extend knowledge of the breadth of involved in many processes in plants, including cytokinins’ activities, we examined effects of natural growth, differentiation and leaf senescence. However, cytokinin bases on the model nematode Caenorhab- they also have various activities in animals. For ditis elegans. We found that kinetin, para-topolin and example, kinetin and trans-zeatin can reduce levels of meta-topolin prolonged the lifespan of C. elegans. several aging markers in human fibroblasts. Kinetin Kinetin also protected the organism against oxidative can also protect mice against oxidative and glyox- and heat stress. Furthermore, our results suggest that idative stress, and prolong fruit flies’ lifespan. presence of reactive oxygen species, but not DAF-16 Additionally, several cytokinins are currently used (the main effector of the insulin/insulin-like growth factor signaling pathway), is required for the beneficial effects of kinetin. Ultra-high performance liquid chromatography-tandem mass spectrometric analysis Electronic supplementary material The online showed that kinetin is unlikely to occur naturally in version of this article (https://doi.org/10.1007/s10522 -017-9742-4) contains supplementary material, which C. elegans, but the worm efficiently absorbs and is available to authorized users. metabolizes it into kinetin riboside and kinetin riboside-5′-monophosphate. A. Kadlecova´ · T. Jirsa · O. Nova´k · M. Strnad · J. Voller (&) Laboratory of Growth Regulators, Centre of the Region Keywords Cytokinin · Kinetin · Topolin · Hana´ for Biotechnological and Agricultural Research, Zeatin · Phytohormones · Aging · Institute of Experimental Botany ASCR, Palacky´ Caenorhabditis elegans University, Sˇlechtitelu˚ 27, 78371 Olomouc, Czech Republic e-mail: [email protected] Introduction J. Kammenga Laboratory of Nematology, Wageningen University, Caenorhabditis elegans (C. elegans) is a free-living Droevendaalsesteeg 1, 6708 PB Wageningen, The Netherlands soil nematode that is used in many research fields, including biogerontology. It is a small, transparent J. Voller animal with a fast life cycle, short lifespan and many Department of Clinical and Molecular Pathology, Institute other advantageous characteristics for biogerontolog- of Molecular and Translational Medicine, Faculty of Medicine and Dentistry, Palacky´ University, Hneˇvotıńska´ ical studies (Corsi 2006). There is significant 3, 77515 Olomouc, Czech Republic homology between C. elegans and human genes, 123 110 Biogerontology (2018) 19:109–120 and many human disease-related genes have coun- inducing adaptive response in lower concentration terparts in worms (Lai et al. 2000). C. elegans has and toxic effect in higher doses. also been used to elucidate genes and pathways that Recently, kinetin has also been shown to directly are involved in aging (Antebi 2007) and conserved influence other cellular processes. For example, it can among various species. The organism is routinely reportedly correct aberrant alternative splicing of pre- used to screen compounds suspected to have lifespan- mRNA in several genetic disorders (Slaugenhaupt prolonging activity, including many substances orig- et al. 2004; Pros et al. 2009), and initial data suggest inating from plants. Such compounds could that kinetin might be effective in patients with potentially be used medically in attempts to prevent familial dysautonomia (Axelrod et al. 2011). Addi- age-associated diseases and increase elderly patients’ tionally, a metabolite of kinetin, kinetin riboside-5′- quality of life. However, effects of cytokinins on the monophosphate, has reported neuroprotective prop- nematode’s lifespan, which could be substantial for erties by acting as a neo-substrate for PINK-1 kinase, reasons outlined below, have not been previously mutations of which are linked to a hereditary form of evaluated. Parkinson’s disease (Hertz et al. 2013). Cytokinins are phytohormones that participate in Trans-zeatin, a cytokinin base consisting of regulation of cell division, differentiation, senescence adenine substituted at the N6-position with an and many other processes in plants (Kieber and isoprenoid side chain, can also retard aging in human Schaller 2014). Interestingly, some cytokinins have fibroblasts (Rattan and Sodagam 2005), reduce the been detected in human cell extracts and urine neurotoxicity of amyloid β in human cell lines, and (Barciszewski et al. 1996, 2000), and animal cells improve the memory of scopolamine-treated mice possess the metabolic pathways required to convert (Choi et al. 2009; Kim et al. 2008). cytokinin bases to ribosides and ribotides (Hertz et al. Both kinetin and trans-zeatin are currently used in 2013; Mlejnek and Kuglik 2000; Mlejnek and cosmetic products, as is a semi-synthetic derivative of Dolezˇel 2005), which also occur in plants. kinetin, pyratine (6-furfuryl-amino-9-(tetrahydropy- Various activities of cytokinins in both mam- ran-2-yl)purine). These products are mostly marketed malian cell cultures and animals have been reported as skin-protective, anti-aging or de-pigmenting (Voller et al. 2017b). The ribosides are often toxic, agents, but lotions containing kinetin and pyratine but cytokinin bases are mostly known to have are also effective in the treatment of photo-damaged protective effects, often ascribed to their antioxidant skin (McCullough et al. 2008; Wanitphakdeedecha activity. Kinetin, a cytokinin base consisting of et al. 2015) and can reduce symptoms of rosacea (Wu adenine substituted at the N6-position with an et al. 2007). There is also some evidence that kinetin aromatic side chain, can protect against oxidative might be effective against psoriasis (An et al. 2017). damage in various systems, both directly by acting as Furthermore, cytokinin ribosides, including kinetin an antioxidant and indirectly by inducing cells’ riboside, have anticancer activity both in vitro and antioxidant defense mechanisms (Olsen et al. 1999; in vivo (Voller et al. 2010). Depending on concen- Sharma et al. 1997; Jabłonska-Trypuc et al. 2016; tration, length of treatment time, and cell line, this Verbeke et al. 2000). For example, McDaniel et al. activity may be mediated by blocking cell cycle (2005) found that kinetin can reduce levels of progression, or by inducing apoptosis either by reactive oxygen species (ROS), lipid peroxidation depleting ATP or disrupting mitochondria (Cabello and DNA damage in cell cultures exposed to UVB. et al. 2009; Cheong et al. 2009; Voller et al. 2017a; Kinetin can also ameliorate several aging markers in Ishii et al. 2002). human cell cultures (Rattan and Clark 1994; Lee In this study, we evaluated effects of kinetin and et al. 2006), prolong the lifespan of fruit flies (Sharma other cytokinin bases on C. elegans, particularly its et al. 1995), and reduce glyoxidative stress both longevity, and probed kinetin’s metabolism and in vitro (Verbeke et al. 2000) and in vivo (in mice mechanism of action in the worm. exposed to galactose) (Liu et al. 2011). Kinetin has also been proposed to act as a hormetin (Rattan 2008), i.e. a compound capable of

123 Biogerontology (2018) 19:109–120 111

Materials and methods was removed once a week for several minutes to provide the worms with sufﬁcient oxygen. Experi- C. elegans strains and maintenance ments were performed under sterile conditions, so no antibiotics or fungicides were required. Experiments Worms were maintained at 20 °C under standard were performed at 25 °C because this results in all of cultivation conditions (Strange 2006). The strains the BA17(fem-1) worms developing into females, used in these experiments were N2, BA17 (fem-1) thereby preventing reproduction and avoiding the and CF1038 (daf-16(mu86)). All strains were pro- need for ﬂuorodeoxyuridine (FuDR) treatment. vided by the Caenorhabditis genetic center (http:// cbs.umn.edu/cgc/home). Stress assays

Chemicals Age-synchronized L4 larvae were washed from the plates, counted and fed with a suspension of E. coli Test compounds, a generous gift from OlChemIm OP50 in S-complete, as in the lifespan experiments. Ltd., included cytokinin bases with either isoprenoid DMSO-treated worms were used as negative controls. (trans-zeatin, dihydrozeatin, cis-zeatin, N6-isopen- Worms were maintained at 20 °C. For oxidative tenyladenine) or aromatic side chains (kinetin, N6- stress assay, worms were exposed to a lethal benzylaminopurine, ortho-topolin, meta-topolin, concentration of juglone (500 µM) 3 days after pre- para-topolin). Dimethylsulfoxide (DMSO) was used treatment, and counted using an inverted microscope as a solvent. All chemicals, including juglone (5- every 1–2 h. To evaluate thermotolerance, worms hydroxy-1,4-naphthoquinone) used for stress assays, were incubated at 35 °C for 90 min, then counted were purchased from Merck. Media and buffers were using an inverted microscope. Animals that failed to prepared as previously described (Solis and Pet- respond to the light stimulus were scored as dead. rascheck 2011). Since N2 (WT) and mutant worm strains were used in the stress assays, the suspensions were supplemented Lifespan experiments with 25 µg/ml FUdR to prevent reproduction.

Lifespan experiments were performed using similar Preparation of worm extracts and cytokinin methodology to Solis and Petrascheck (2011), with analysis by UHPLC-MS/MS several modifications. Briefly, BA17(fem-1) worms were bleached and age-synchronized progeny were Worms were exposed to either kinetin or vehiculum cultivated at 25 °C until they reached adulthood. alone in liquid S-complete medium supplemented Young adults were washed from NGM plates with with living or heat-killed bacteria (80 °C for 15 min). S-complete medium and counted, then the resulting After 24 h, worms were washed four times with M9 suspension was diluted with S-complete. Worms buffer and snap-frozen in liquid nitrogen. Escherichia were fed with a suspension of Escherichia coli OP50 coli strain OP50, which was used as a food source, in S-complete, then treated with the compounds and was cultivated overnight in LB medium containing pipetted into 96-well plates (approximately 15 indi- either kinetin or DMSO. Bacteria were centrifuged viduals per well). Plates were sealed to prevent and the pellets were washed with sterile water. Pellets evaporation of the medium and kept at 25 °C. Living were then re-suspended in M9 buffer and centrifuged. worms were identified based on their movement, and This process was repeated twice. The supernatant was counted using an inverted microscope three times a removed, wet pellets were weighed, and samples week. Movement was induced by a light stimulus were snap-frozen in liquid nitrogen. All samples were from the microscope and by briefly shaking the plate stored at − 70 °C prior to testing. on a plate shaker prior to counting. Worms that failed Cytokinin metabolites were quantified as previ- to finish normal development, worms that were ously described (Svacˇinova´ et al. 2012), with injured by pipetting during preparation of the exper- modifications described by Antoniadi et al. (2015). iment and groups of worms permanently tangled Samples (1 ml) were homogenized and extracted in together were excluded from the analysis. The seal 1 ml of 80% methanol (MeOH) containing a cocktail 123 112 Biogerontology (2018) 19:109–120 of stable isotope-labeled internal standards (0.25 pmol Results of cytokinin bases and ribosides, and 0.5 pmol of cytokinin nucleotides per sample). The extracts were The effect of cytokinin bases on the worms’ purified using an Oasis MCX column (60 mg/3 ml, lifespan Waters) conditioned with 1 ml each of 100% MeOH and H2O, then equilibrated sequentially with 1 ml of Test compounds included cytokinins with both aro- 50% (v/v) nitric acid, 1 ml of H2O, and 1 ml of 1 M matic (kinetin, N6-benzylaminopurine, ortho-topolin, HCOOH. After samples had been applied to the MCX meta-topolin, para-topolin) and isoprenoid (N6- column, unretained compounds were removed by isopentenyladenine, trans-zeatin, dihydrozeatin, cis- washing with 1 ml of HCOOH (1 M) and 1 ml 100% zeatin) side chains. All compounds were tested at MeOH. Pre-concentrated analytes were eluted in two three concentrations—200, 100 and 50 µM—except µ steps, using 1 ml of 0.35 M NH4OH aqueous solution topolins, which precipitated in the medium at 200 M followed by 2 ml of 0.35 M NH4OH in 60% (v/v) and so were not tested at this concentration. None of MeOH solution. The eluates were then evaporated to the tested compounds were acutely toxic to C. dryness in vacuo and stored at − 20 °C. Cytokinin elegans and treated worms displayed no behavioral levels were determined using ultra-high performance or morphological abnormalities compared to control liquid chromatography-electrospray tandem mass worms. Kinetin reportedly retards development in spectrometry (UHPLC-MS/MS) with stable isotope- fruit flies (Sharma et al. 1995), but we observed no labelled internal standards as references. Three such effect in C. elegans: growth rates of worms independent biological replicates were analyzed. cultivated in kinetin for three generations were Concentrations were initially measured in pmol/ indistinguishable from controls cultivated in vehicle 100 worms and pmol/g of bacterial wet pellet. They alone (data not shown). were then converted to µM, assuming that each worm Four compounds—kinetin, trans-zeatin, meta- had the average volume of 9.54 pl reported by So topolin and para-topolin—were initially identified et al. (2011), and the E. coli cells’ density was that significantly increased the worms’ lifespan 1.09 mg/ml, as they consist of 2/3 water and 1/3 other (Table 1 and Supplementary Materials Table 1). components, e.g. protein, with a typical density of Compounds showing significant activity in the 1.3 mg/ml, as reported in the BioNumbers database initial screening were re-tested at the concentration at (Milo et al. 2009). which they showed the greatest efficacy. Repeated experiments confirmed the ability of kinetin, para- Statistical analysis topolin and meta-topolin to prolong worm lifespan. Despite trans-zeatin yielding promising results in the Descriptive statistics—mean, standard deviation, initial screening, the effects were much weaker and median and 90th quantile (Q90) survival—and insignificant in two repeated experiments with larger Kaplan–Meier survival curves were calculated using experimental populations (Table 2, Supplementary the OASIS 2 online application (Han et al. 2016). The Materials Table 2). Figures 1, 2, and 3 show Kaplan– log-rank test was used to compare survival curves of Meier survival curves for worms treated with kinetin, control populations with those of populations treated meta-topolin, and para-topolin, respectively. with cytokinins. p values for each comparison in an We further characterized the protective activity of experiment were corrected using the Bonferroni kinetin to facilitate the interpretation of any observed method. A two-tailed Z-test for proportions was effects in C. elegans in the context of findings by applied (in R) when comparing effects of compounds other researchers, as it is the most intensively studied on survival at a single time-point. cytokinin known to have such activity.

Kinetin increases thermotolerance and resistance to oxidative stress

Next, as manipulations that prolong worms’ lifespan also often increase their stress resistance (Zhou et al. 123 Biogerontology (2018) 19:109–120 113

Table 1 Results of initial screenings of natural cytokinin bases’ effects on C. elegans lifespan Compound Concentration (µM) N Mean—% change Median—% change Q90—% change P value kinetin 200 143 + 21 + 20 + 18 5.6 9 10−7 100 121 + 6 + 5 + 4 [ 0.05 50 75 + 2 + 5 +2 [ 0.05 N6-isopentenyladenine 200 107 − 2 − 6 − 1 [ 0.05 100 108 − 2 + 2 − 1 [ 0.05 50 70 − 8 − 4 − 6 [ 0.05 N6-benzyladenine 200 111 + 1 + 2 − 2 [ 0.05 100 100 + 3 + 50[ 0.05 50 91 0 − 17 + 6 [ 0.05 ortho-topolin 100 138 + 3 + 2 + 15 [ 0.05 50 132 + 6 + 5 + 20 [ 0.05 meta-topolin 100 108 0 − 3 + 10 [ 0.05 50 130 + 10 + 8 + 27 1.2 9 10−2 para-topolin 100 143 + 8 + 8 + 21 1.2 9 10−2 50 111 + 5 + 1 + 18 [ 0.05 trans-zeatin 200 96 − 9 − 15 − 7 [ 0.05 100 115 + 1 − 5 + 2 [ 0.05 50 51 + 17 + 19 + 15 1.2 9 10−4 cis-zeatin 200 122 − 4 − 8 − 2 [ 0.05 100 133 − 3 − 5 − 6 [ 0.05 50 129 − 2 − 4 + 6 [ 0.05 dihydrozeatin 200 117 − 6 − 10 − 5 [ 0.05 100 126 − 1 − 50[ 0.05 50 131 0 − 40[ 0.05 The table shows % change of mean (average) and median (time point at which 50% of the population had died) lifespan and Q90 of populations treated with the indicated compounds in comparison to control populations. Q90—the time point at which 90% of worms die—was selected as a more robust indicator of such effects than maximal lifespan (time at which the last worm dies). N indicates the size of the population. Average N for the control population was 105

Table 2 Results of conﬁrmatory experiments with four active natural cytokinin bases selected from initial screening experiments Compound Concentration (µM) Mean—% change N Median—% change Q90—% change P value kinetin 200 + 6 113 + 5 + 639 10−2 meta-topolin 50 + 15 113 + 20 + 16 1.7 9 10−2 para-topolin 100 + 12 128 + 10 + 12 4.6 9 10−2 trans-zeatin 50 − 3 125 − 8 − 4 [ 0.05 + 2 104 + 4 + 1 [ 0.05 The table shows % change of mean (average) and median (time point at which 50% of the population had died) lifespan and Q90 of populations treated with the indicated compounds in comparison to control populations. N indicates the size of the population. Average N for the control population was 107

2011), we tested the possibility that the longevity- involving exposing worms to 100 or 200 µMof enhancing effect of kinetin is linked to increases in kinetin. This pre-treatment protected the worms from resistance to various stresses, in further assays oxidative stress caused by exposure to a lethal 123 114 Biogerontology (2018) 19:109–120

Fig. 1 Effect of 200 µM kinetin on aging of a C. elegans Fig. 4 Effect of kinetin on survival of worms exposed to population juglone. Pre-treatment with kinetin caused a signiﬁcant increase in survival (p = 2.8 9 10−4 and 1.8 9 10−2 for kinetin 200 µM and 100 µM respectively). Sizes of populations treated with 100 µM kinetin, 200 µM kinetin and controls were n = 467, 421 and 470, respectively

average increase of 14% in the survival of kinetin (200 µM)-treated worms compared to control worms (p \ 0.05 in most cases). The increase was smaller (average increase of 5%) and insigniﬁcant in most cases for populations pre-treated with 100 µMof kinetin. These experiments were repeated four times (for overall results see Supplementary Materials Fig. 2 Effect of 50 µM meta-topolin on aging of a C. elegans Table 6). population The effect of kinetin is dependent on ROS

It was previously proposed that the beneficial effect of kinetin is a result of its antioxidant activity. However, co-treating worms with 200 µM kinetin and the antioxidant trolox (100 µM) significantly reduced kinetin’s ability to protect worms against oxidative stress induced by juglone exposure. Repre- sentative results are shown in Fig. 5. This effect was observed in three independent experiments (Supple- mentary Materials Table 5). Similarly, the thermotolerance of populations pre- Fig. 3 Effect of 100 µM para-topolin on aging of a C. elegans µ + µ population treated with kinetin (200 M) trolox (100 M) was reduced. After exposure to heat stress (35 °C for concentration of the ROS generator juglone. In four 90 min), the average survival was 15% lower than for \ independent experiments, 200 µM of kinetin signif- worms treated with kinetin alone (p 0.05) and even [ icantly increased the survival of worms (p \ 0.05), 5% lower than for untreated control worms (p 0.05 while the effect of 100 µM kinetin was weaker and in most cases). This experiment was repeated three only significant in some cases (Supplementary Mate- times (Supplementary Materials Table 7). rials Table 4). Representative Kaplan–Meier curves We also decided to perform one lifespan experi- are shown in Fig. 4. ment to see if the suppressive effect of trolox was Kinetin also increased the thermotolerance of C. reproducible also in these experimental settings. elegans. Exposure to 35 °C for 90 min resulted in an Indeed, also in the lifespan experiment the effect of 123 Biogerontology (2018) 19:109–120 115

population, an increase in survival of borderline significance was observed (Supplementary Materials Table 4). In the second experiment, with a larger population, kinetin significantly prolonged the survival of daf-16 mutants at both 100 and 200 µM (Fig. 7). Kinetin also increased the thermotolerance of these mutants. After exposure to 35 °C for 90 min, the survival of populations treated with 200 µM kinetin was, on average, 15% higher than in control populations (p \ 0.05). The effect of 100 µM kinetin Fig. 5 Survival of worms pre-treated with kinetin and trolox was smaller, a 10% increase in survival on average, after juglone exposure. Kinetin-treated worms showed significantly higher survival rates than both control worms and not significant in two out of three experiments (p = 3 9 10−3) and trolox co-treated worms (p = 3.6 9 (Supplementary Materials Table 6). −2 10 ). Sizes of populations treated with 200 µM kinetin, Our results suggest that the effect of kinetin is µ = = 200 M kinetin trolox and controls were n 109, 129 and independent of DAF-16, the major effector of Insulin 182, respectively and Insulin-like growth factor (IGF) Signaling (IIS). kinetin was suppressed by trolox (Fig. 6 and Supple- mentary Table 3). C. elegans worms absorb and ribosylate kinetin Together, these results suggest that the longevity- effectively enhancing effect of kinetin in worms is probably not based on its ability to scavenge ROS, and that ROS As already mentioned, kinetin is present in other must be present for this effect of kinetin. organisms besides plants, including humans. Thus, we also compared levels of endogenous kinetin in DAF-16 is not required for kinetin-induced stress control and kinetin-treated populations of C. elegans. resistance Moreover, as cytokinin bases can be metabolized into the corresponding ribosides and riboside-5′-phos- DAF-16 is a transcription factor that participates in phates in both mammals and plants, to varying regulation of various genes involved in stress degrees, we also examined levels of kinetin riboside ′ responses and lifespan extension (Tullet 2015). Thus, (KR) and kinetin riboside-5 -monophosphate (KMP) we evaluated the possibility that DAF-16 may in these populations. The compounds were analyzed mediate effects of kinetin in C. elegans, by assaying in two sets of replicate samples (each consisting of kinetin’s ability to increase the oxidative stress ca. 1000 worms) using UHPLC-MS/MS. Neither resistance of daf-16(mu86) mutants exposed to juglone. This experiment was repeated twice. In the first experiment, using 200 µM kinetin and a small

Fig. 7 Survival of daf-16 null mutants pre-treated with kinetin after juglone exposure. Kinetin had signiﬁcant effects at both 200 μM(p= 3.9 9 10−5) and 100 µM(p= 6.5 9 10−6). Sizes Fig. 6 Comparison of effects of 200 µM kinetin with and of populations treated with 200 µM kinetin, 100 µM kinetin without 100 µM trolox on aging of C. elegans populations and controls were n = 368, 336 and 265, respectively 123 116 Biogerontology (2018) 19:109–120

Table 3 Concentrations (µM) of kinetin and its metabolites in extracts from worms and bacteria after exposure to 100 µM kinetin for 24 h Sample type Kinetin Kinetin riboside Kinetin riboside-5′-monophosphate

C. elegans, standard cultivation conditions 160.34 17.58 372.59 970.14 71.5 508.23 C. elegans, heat-killed bacteria 33.83 9.47 30.51 80.55 195.15 127.61 E. coli OP50 0.54 0.31 ND 0.28 0.22 ND

kinetin nor its metabolites were detected in control other naturally occurring cytokinin bases, some of samples, suggesting that the worms do not naturally which have not previously been tested for anti-aging produce these compounds (Table 3). However, in and cyto-protective activities, in the model organism extracts of worms pre-treated with kinetin (100 µM, C. elegans. 24 h) grown under standard cultivation conditions, We found that kinetin, and two aromatic cytoki- the parent compound was detected at levels exceed- nins, meta-topolin and para-topolin, can increase ing the applied concentration, demonstrating that it is worms’ longevity. Therefore, our data show that the absorbed very efficiently. We also detected both KR lifespan-prolonging activity of kinetin is not limited and KMP in these worms, indicating that they to insects. It would be interesting to see whether its metabolize kinetin. There was significant variation effect on lifespan extends to other model vertebrates in both the concentrations and ratios of the three such as fish and rodents. While para-topolin is known metabolites between the two sets of samples, which to have antioxidant activity in vitro (Brizzolari et al. can be ascribed to biological variability and the loss 2016), there is limited data regarding the cyto- of some worms during washing steps. protective activity of topolins in cell cultures and To ensure that the detected metabolites were animals. The activity we observed suggests that produced by the worms and not by the E. coli OP50 topolins warrant greater attention in future studies. A bacteria used as food, we tested for the presence of caveat is that the longevity-promoting effect was the metabolites in these bacteria, and in worms fed typically only observed at the highest tested concen- with heat-killed bacteria. KR, but not KMP, was trations, and these compounds’ low solubility detected in the bacteria, demonstrating that KMP is prohibited testing at higher concentrations. not produced by bacteria. Both KR and KMP were Another cytokinin that has previously been exten- detected in worms fed with heat-killed bacteria. sively studied in connection with aging and age- related diseases is trans-zeatin. It has youth-preserving activity in cell cultures (Rattan and Sodagam Discussion 2005) and neuroprotective effects in both cell cultures and rodents (Choi et al. 2009; Kim et al. Cytokinins have various protective activities both 2008). In our study, this compound showed promis- in vitro and in vivo. One compound from this group, ing results in an initial experiment, but its effect was kinetin, has known ability to prolong the lifespan of not significant in repeated experiments using larger fruit flies (Sharma et al. 1995). Moreover, kinetin can populations. reduce oxidative stress (Olsen et al. 1999) and After initial experiments we focused on kinetin. glyoxidative stress both in vitro (Verbeke et al. Kinetin increased the resistance of C. elegans to 2000) and in vivo (Liu et al. 2011). In cell culture, oxidative stress caused by juglone, which generates kinetin can also ameliorate several markers of aging superoxide anions by redox cycling (Inbaraj and (Rattan and Clark 1994; Lee et al. 2006). In this Chignell 2004), and increased their thermotolerance. study, we evaluate the protective effect of kinetin and

123 Biogerontology (2018) 19:109–120 117

We also performed several experiments to probe the Kinetin has been previously reportedly identified mechanism of kinetin’s protective activity in worms. in DNA extracted from human cells (Barciszewski First, we tested possible involvement of the et al. 1996) and urine from cancer patients (Bar- insulin/insulin-like growth factor signaling pathway ciszewski et al. 2000), suggesting that it can occur (IIS), as mutations in IIS components affect the naturally under some circumstances in other organ- lifespan of various species (Barbieri et al. 2003), and isms besides plants. Using UHPLC-MS/MS, we many compounds that can prolong the lifespan and measured levels of endogenous kinetin in worms increase stress resistance in worms affect the IIS. A and compared them to its levels in a kinetin-treated key effector of IIS is DAF-16/FOXO (Forkhead box population. We found neither kinetin nor its deriva- protein O), a transcription factor that activates many tives in control worms, indicating that these genes involved in stress responses or dauer larvae compounds are not normally produced by C. elegans, formation (Tullet 2015). Reductions in IIS pathway at least under our cultivation conditions. activity result in translocation of DAF-16 into the However, exogenously applied kinetin (100 µM, nucleus and trans-activation of its target genes. 24 h) was efficiently absorbed by the worms, to levels Whereas worms with mutations in the daf-2 gene exceeding the applied concentration. In addition to (encoding the C. elegans insulin-like growth factor 1 kinetin, we also detected kinetin riboside (KR) and receptor homologue DAF-2) have significantly kinetin riboside-5′-monophosphate (KMP). increased lifespans and stress resistance (Kenyon The presence of kinetin metabolites at such high et al. 1993; Lithgow et al. 1995), mutants with concentrations raises the interesting possibility that impaired DAF-16 have shorter lifespans than WT these metabolites are responsible for the beneficial worms. Therefore, we tested effects of kinetin in daf- effects of kinetin treatment. We propose the follow- 16 null mutants, and found that it increased their ing hypotheses regarding these metabolites’ stress resistance as well, showing that IIS activity is contributions to the protective activity of kinetin not essential for kinetin’s activity. based on their effects on human cells. In mammalian ROS have been traditionally regarded as important cells, cytokinin ribosides, including KR, are highly pro-aging factors (the free radical theory of aging), cytotoxic as they induce ATP depletion (Ishii et al. mostly due to their ability to directly damage cells’ 2002; Cabello et al. 2009), possibly by impairing macromolecules (Harman 1955). Recent studies sug- mitochondrial functions (Cheong et al. 2009). How- gest that their role in aging is more complex than this, ever, at low concentrations, two cytokinin ribosides, but certainly no less important. For example, they can N6-isopentenyladenosine and N6-benzyladenosine, act as signaling molecules in many pathways, directly reportedly activate NRF-2 (Dassano et al. 2014), modify certain proteins involved in signal transduc- suggesting they may protect cells from oxidative tion, and may participate in the epigenetic stress. This raises the possibility that these com- modulation of gene expression (Davalli et al. 2016). pounds act as hormetins—compounds that induce ROS are also involved in many interventions that hormetic adaptive mechanisms. We showed that the increase longevity and they activate hormetic adap- beneficial effect of kinetin is suppressed by the tive responses—a process sometimes referred to as presence of an antioxidant. Perturbation of mitochon- mitochondrial hormesis, or mitohormesis (Ristow and dria in C. elegans by KR, as in human cells, would Schmeisser 2014). lead to the production of ROS and possibly activation The cyto-protective effects of cytokinins are often of an adaptive response. Therefore, we speculate that ascribed to their intrinsic antioxidant activities. kinetin might act as a hormetin precursor, a pro- However, we found that co-treatment with the known hormetin. antioxidant trolox (an analog of vitamin E) signifi- Moreover, it was recently hypothesized that effects cantly reduced or even completely suppressed the of another cytokinin riboside, N6-isopenteny- effects of kinetin. Given the ROS-scavenging activity ladenosine (iPR), could be mediated by activation of trolox, we hypothesize that kinetin activity of AMP-activated protein kinase (AMPK) by its requires the presence of ROS. However, the exact metabolite iPR-5′-monophosphate in human cells role of ROS in the beneficial effects of kinetin (Pisanti et al. 2014). AMPK senses the ratio of remains to be determined. ATP/ADP/AMP in cells, thereby providing 123 118 Biogerontology (2018) 19:109–120 information about cellular energy levels (Hardie and role, possibly via activation of adaptive responses Hawley 2001). The enzyme also plays an important and/or activation of AMP-activated protein kinase role in aging, as its over-expression (Apfeld et al. (AMPK). 2004), or activation by compounds such as metformin (Onken and Driscoll 2010), increases longevity in Acknowledgements Strains used in this study were provided worms. It is tempting to speculate that KMP might by the CGC, which is funded by the NIH Office of Research ′ Infrastructure Programs (P40 OD010440). The authors are have a similar effect to iPR-5 -monophosphate. High grateful to Hana Martıńkova´ for her help with phytohormone concentrations of KMP could also activate AMPK in analyses. This study was supported by the Ministry of worms, since the enzyme is highly conserved across Education, Youth and Sports of the Czech Republic phylla (Hardie and Hawley 2001). AMPK activation (National Program for Sustainability I, grant nos. LO1204 and LO1304; INTER-COST LTC17 project code LTC17072). induces respiration and catabolism (Schulz et al. This article is based upon work from COST Action BM1408, 2007), thereby increasing ROS production which can supported by COST (European Cooperation in Science and be suppressed by antioxidants (Schulz et al. 2007; Technology). Mungai et al. 2011; De Haes et al. 2014). This would be consistent with our findings that ROS are required for the effects of kinetin. References Effects of exogenously applied kinetin metabolites An S, Cha HJ, Ko JM, Han H, Kim SY, Kim KS, Lee SJ, An on the lifespan of C. elegans, and tests of the IS, Kim S, Youn HJ et al (2017) Kinetin improves barrier hypotheses outlined above, will be included in function of the skin by modulating keratinocyte differ- planned studies. Future studies will also include entiation markers. Ann Dermatol 29(1):6–12 further analysis of kinetin´s metabolism, for example Antebi A (2007) Genetics of aging in Caenorhabditis elegans. PLoS Genet 3(9):e129 to establish whether C. elegans produces higher order Antoniadi I, Placˇkova´ L, Simonovik B, Dolezˇal K, Turnbull C, phosphates, which also have beneficial activity in Ljung K, Nova´k O (2015) Cell-type-specific cytokinin human cells (Hertz et al. 2013). distribution within the Arabidopsis primary root apex. Plant Cell 27(7):1955–1967 Apfeld J, O’Connor G, McDonagh T, DiStefano P, Curtis R (2004) The AMP-activated protein kinase AAK-2 links Conclusion energy levels and insulin-like signals to lifespan in C. elegans. Genes Dev 18(24):3004–3009 The plant hormone kinetin can prolong the lifespan of Axelrod F, Liebes L, Gold-von Simson G, Mendoza S, Mull J, Leyne M, Norcliffe-Kaufmann L, Kaufmann H, Slau- fruit flies and ameliorate aging markers in human cell genhaupt S (2011) Kinetin improves IKBKAP mRNA cultures. Here, we report that kinetin can prolong the splicing in patients with familial dysautonomia. Pediatr lifespan of another model organism, C. elegans.We Res 70(5):480–483 also report the longevity-enhancing effect of two Barbieri M, Bonafe` M, Franceschi C, Paolisso G (2003) Insulin/IGF signaling pathway: an evolutionarily con- other cytokinin bases, meta-topolin and para-topolin, served mechanism of longevity from yeast to humans. Am for the first time. In addition to prolonging lifespan, J Physiol 285(5):1064–1071 kinetin increased the worms’ stress resistance. Its Barciszewski J, Siboska G, Pedersen B, Clark B, Rattan S effect was independent of the transcription factor (1996) Evidence for the presence of kinetin in DNA and cell extracts. FEBS Lett 393(2):197–200 DAF-16, and thus is not probably connected to the Barciszewski J, Mielcarek M, Stobiecki M, Siboska G, Clark B insulin/insulin-like growth factor pathway. However, (2000) Identification of 6-furfuryladenine (kinetin) in addition of an antioxidant significantly reduced its human urine. Biochem Biophys Res Commun 279(1):69– beneficial effect, indicating a crucial role of ROS. We 73 Brizzolari A, Marinello C, Carini M, Santaniello E, Biondi P also analyzed endogenous levels of kinetin in C. (2016) Evaluation of the antioxidant activity and capacity elegans. Our results show that kinetin does not occur of some natural N6-substituted adenine derivatives (cy- naturally in the organism, but it is effectively tokinins) by fluorimetric and spectrophotometric assays. absorbed after exogenous exposure and, as in human J Chromatogr B 1019:164–168 Cabello C, Bair W, Ley S, Lamore S, Azimian S, Wondrak G cells, it is metabolized to kinetin riboside and kinetin (2009) The experimental chemotherapeutic N6-furfury- riboside-5′-monophosphate. Kinetin’s exact mecha- ladenosine (kinetin-riboside) induces rapid ATP nism of action in C. elegans remains unclear, but we depletion, genotoxic stress, and CDKN1A (p21) hypothesize that its metabolites play an important 123 Biogerontology (2018) 19:109–120 119

upregulation in human cancer cell lines. Biochem Phar- Caenorhabditis elegans by comparative proteomics. macol 77(7):1125–1138 Genome Res 10(5):703–713 Cheong J, Goh D, Yong J, Tan S, Ong E (2009) Inhibitory Lee JH, Chung KY, Bang D, Lee KH (2006) Searching for effect of kinetin riboside in human heptamoa, HepG2. aging-related proteins in human dermal microvascular Mol BioSyst 5(1):91–98 endothelial cells treated with anti-aging agents. Pro- Choi S, Jeong C, Choi S, Chun J, Kim Y, Lee J, Shin D, Heo H teomics 6(4):1351–1361 (2009) Zeatin prevents amyloid β-induced neurotoxicity Lithgow G, White T, Melov S, Johnson T (1995) Thermotol- and scopolamine-induced cognitive deficits. J Med Food erance and extended life-span conferred by single-gene 12(2):271–277 mutations and induced by thermal stress. Proc Natl Acad Corsi A (2006) A biochemist’s guide to C. elegans. Anal Sci 92(16):7540–7544 Biochem 359(1):1 Liu Y, Zhang Z, Yang X (2011) Kinetin protects against lipid Dassano A, Mancuso M, Giardullo P, De Cecco L, Ciuffreda P, peroxidation and improves antioxidant status in cultured Santaniello E, Saran A, Dragani TA, Colombo F (2014) astrocytes and mouse brain exposed to D-galactose. Afr J N6-isopentenyladenosine and analogs activate the NRF2- Biotech 10(55):11721–11727 mediated antioxidant response. Redox Biol 2:580–589 Mattson M (2008) Hormesis defined. Ageing Res Rev 7(1):1–7 Davalli P, Mitic T, Caporali A, Lauriola A, D’Arca D (2016) McCullough J, Garcia R, Reece B (2008) A clinical study of ROS, cell senescence, and novel molecular mechanisms in topical Pyratine 6 for improving the appearance of pho- aging and age-related diseases. Oxidative Med Cell todamaged skin. J Drugs Dermatol 7(2):131–135 Longev 2016:1–18 McDaniel D, Neudecker B, DiNardo J, Lewis J, Maibach H De Haes W, Frooninckx L, Van Assche R, Smolders A, (2005) Idebenone: a new antioxidant–part 1. Relative Depuydt G, Billen J, Braeckman B, Schoofs L, Tem- assessment of oxidative stress protection capacity com- merman L (2014) Metformin promotes lifespan through pared to commonly known antioxidants. J Cosmet mitohormesis via the peroxiredoxin PRDX-2. Proc Natl Dermatol 4(1):10–17 Acad Sci 111(24):E2501–E2509 Milo R, Jorgensen P, Moran U, Weber G, Springer M (2009) Han S, Lee D, Lee H, Kim D, Son H, Yang J, Lee S, Kim S BioNumbers—the database of key numbers in molecular (2016) OASIS 2: online application for survival analysis 2 and cell biology. Nucleic Acids Res 38(suppl_1):D750– with features for the analysis of maximal lifespan and D753 healthspan in aging research. Oncotarget 7(35):56147– Mlejnek P, Dolezˇel P (2005) Apoptosis induced by N6-sub- 56152 stituted derivatives of adenosine is related to intracellular Hardie D, Hawley S (2001) AMP-activated protein kinase: the accumulation of corresponding mononucleotides in HL- energy charge hypothesis revisited. Bioes- says 23 60 cells. Toxicol In Vitro 19(7):985–990 (12):1112–1119 Mlejnek P, Kuglik P (2000) Induction of apoptosis in HL-60 Harman D (1955) Aging: a theory based on free radical and cells by N6-benzyladenosine. J Cell Biochem 77(1):6–17 radiation chemistry. J Gerontol 11(3):298–300 Mungai P, Waypa G, Jairaman A, Prakriya M, Dokic D, Ball Hertz N, Berthet A, Sos M, Thorn K, Burlingame A, Nakamura M, Schumacker P (2011) Hypoxia triggers AMPK acti- K, Shokat K (2013) A neo-substrate that amplifies cat- vation through reactive oxygen species-mediated alytic activity of Parkinson’s-disease-related kinase activation of calcium release-activated calcium channels. PINK1. Cell 154(4):737–747 Mol Cell Biol 31(17):3531–3545 Inbaraj JJ, Chignell CF (2004) Cytotoxic action of juglone and Olsen A, Siboska G, Clark B, Rattan S (1999) N6-furfury- plumbagin: a mechanistic study using HaCaT ker- ladenine, kinetin, protects against Fenton reaction- atinocytes. Chem Res Toxicol 17(1):55–62 mediated oxidative damage to DNA. Biochem Biophys Ishii Y, Hori Y, Sakai S, Honma Y (2002) Control of differ- Res Commun 265(2):499–502 entiation and apoptosis of human myeloid leukemia cells Onken B, Driscoll M (2010) Metformin induces a dietary by cytokinins and cytokinin nucleosides, plant rediffer- restriction–like state and the oxidative stress response to entiation-inducing hormones. Cell Growth Differ 13 extend C. elegans healthspan via AMPK, LKB1, and (1):19–26 SKN-1. PLoS ONE 5(1):e8758 Jabłonska-Trypuc´ A, Matejczyk M, Czerpak R (2016) N6- Pisanti S, Picardi P, Ciaglia E, Margarucci L, Ronca R, Gia- benzyladenine and kinetin influence antioxidative stress comini A, Malfitano AM, Casapullo A, Laezza C, parameters in human skin fibroblasts. Mol Cell Biochem Gazzerro P et al (2014) Antiangiogenic effects of N6- 413(12):97–107 isopentenyladenosine, an endogenous isoprenoid end Kenyon C, Chang J, Gensch E, Rudner A, Tabtiang R et al product, mediated by AMPK activation. FASEB J 28 (1993) A C. elegans mutant that lives twice as long as (3):1132–1144 wild type. Nature 366(6454):461–464 Pros E, Fernańdez-Rodrı´guez J, Benito L, Ravella A, Capella´ Kieber J, Schaller G (2014) Cytokinins. Arabidopsis Book 12: G, Blanco I, Serra E, La´zaro C (2009) Modulation of e0168 aberrant NF1 pre-mRNA splicing by kinetin treatment. Kim M, Choi S, Lim S, Kim H, Kim Y, Yoon H, Shin D (2008) Eur J Hum Genet 18(5):614–617 Zeatin supplement improves scopolamine- induced Rattan S (2008) Principles and practice of hormetic treatment memory impairment in mice. Biosci Biotechnol Biochem of aging and age-related diseases. Hum Exp Toxicol 27 72(2):577–581 (2):151–154 Lai C, Chou C, Ch’ang L, Liu C, Lin W (2000) Identification of novel human genes evolutionarily conserved in 123 120 Biogerontology (2018) 19:109–120

Rattan S, Clark B (1994) Kinetin delays the onset of aging isolation from Arabidopsis tissues using miniaturized characteristics in human fibroblasts. Biochem Biophys purification: pipette tip solid-phase extraction. Plant Res Commun 201(2):665–672 Methods 8(1):17 Rattan S, Sodagam L (2005) Gerontomodulatory and youth- Tullet J (2015) DAF-16 target identification in C. elegans: past, preserving effects of zeatin on human skin fibroblasts present and future. Biogerontology 16(2):221–234 undergoing aging in vitro. Rejuvenation Res 8(1):46–57 Verbeke P, Siboska G, Clark B, Rattan S (2000) Kinetin Ristow M, Schmeisser K (2014) Mitohormesis: promoting inhibits protein oxidation and glycoxidation in vitro. health and lifespan by increased levels of reactive oxygen Biochem Biophys Res Commun 276(3):1265–1270 species (ROS). Dose Response 12(2):288–341 Voller J, Zatloukal M, Lenobel R, Dolezˇal K, Be´resˇ T, Krysˇtof Schulz T, Zarse K, Voigt A, Urban N, Birringer M, Ristow M V, Spıćhal L, Niemann P, Dzˇuba´k P, Hajdućh M (2010) (2007) Glucose restriction extends Caenorhabditis ele- Anticancer activity of natural cytokinins: a structure–ac- gans life span by inducing mitochondrial respiration and tivity relationship study. Phytochemistry 71(11):1350– increasing oxidative stress. Cell Metab 6(4):280–293 1359 Sharma S, Kaur P, Rattan S (1995) Plant growth hormone Voller J, Be´res T, Zatloukal M, Kaminski P, Niemann P, kinetin delays aging, prolongs the lifespan, and slows Dolezˇal K, Dzˇuba´k P, Hajdućh M, Strnad M (2017a) The down development of the fruitfly Zaprionus paravittiger. natural cytokinin 2OH3MeOBAR induces cell death by a Biochem Biophys Res Commun 216(3):1067–1071 mechanism that is different from that of the “classical” Sharma S, Kaur J, Rattan S (1997) Increased longevity of cytokinin ribosides. Phytochemistry 136:156–164 kinetin-fed Zaprionus fruitflies is accompanied by their Voller J, Makova´ B, Kadlecova´ A, Gonzalez G, Strnad M reduced fecundity and enhanced catalase activity. IUBMB (2017b) Plant hormone cytokinins for modulating human Life 41(5):869–875 aging and age-related diseases. In: Hormones in ageing Slaugenhaupt S, Mull J, Leyne M, Cuajungco M, Gill S, Hims and longevity. Springer, pp 311–335 M, Quintero F, Axelrod F, Gusella J (2004) Rescue of a Wanitphakdeedecha R, Meeprathom W, Manuskiatti W et al human mRNA splicing defect by the plant cytokinin (2015) Efficacy and safety of 0.1% kinetin cream in the kinetin. Hum Mol Genet 13(4):429–436 treatment of photoaging skin. Indian J Dermatol Venereol So S, Miyahara K, Ohshima Y (2011) Control of body size in Leprol 81(5):547 C. elegans dependent on food and insulin/IGF-1 signal. Wu J, Weinstein G, Kricorian G, Kormeili T, McCullough J Genes Cells 16(6):639–651 (2007) Topical kinetin 0.1% lotion for improving the Solis G, Petrascheck M (2011) Measuring Caenorhabditis signs and symptoms of rosacea. Clin Exp Dermatol 32 elegans life span in 96 well microtiter plates. J Vis Exp. (6):693–695 https://doi.org/10.3791/2496 Zhou K, Pincus Z, Slack F (2011) Longevity and stress in Strange K (2006) C. elegans: methods and applications book Caenorhabditis elegans. Aging (Albany NY) 3(8):733– (series: methods in molecular biology) 753 Svacˇinova´ J, Nova´k O, Placˇkova´ L, Lenobel R, Holı´k J, Strnad M, Dolezˇal K (2012) A new approach for cytokinin

123 7.2 Supplement 2

Kadlecová, A., Maková, B., Artal-Sanz, M., Strnad, M. & Voller, J. (2019). The plant hormone kinetin in disease therapy and healthy aging. Ageing Research Reviews, in press.

56 Ageing Research Reviews xxx (xxxx) xxxx

Contents lists available at ScienceDirect

Ageing Research Reviews

journal homepage: www.elsevier.com/locate/arr

Review The plant hormone kinetin in disease therapy and healthy aging ⁎ Alena Kadlecováa, Barbara Makováa, Marta Artal-Sanzc, Miroslav Strnada,Jiří Vollera,b, a Laboratory of Growth Regulators, The Czech Academy of Sciences, Institute of Experimental Botany & Palacký University, Šlechtitelů 27, CZ-78371 Olomouc, Czech Republic b Department of Clinical and Molecular Pathology, Institute of Molecular and Translational Medicine, Faculty of Medicine and Dentistry, Palacký University, Hněvotínská 3, 775 15 Olomouc, Czech Republic c Andalusian Centre for Developmental Biology, CISIC-JA-University Pablo de Olavide, Department of Molecular Biology and Biochemical Engineering, Carretera de Utrera km 1, 41013 Sevilla, Spain

ARTICLE INFO ABSTRACT

Keywords: It has been more than 60 years since the discovery of kinetin, the ﬁrst known member of a group of plant Cytokinin hormones called cytokinins. In this review we summarize the health-promoting activity of kinetin in animal Kinetin systems, ranging from cells cultured in vitro through invertebrates to mammals. Kinetin has been shown to Aging modulate aging, to delay age-related physiological decline and to protect against some neurodegenerative dis- Age-related diseases eases. We also review studies on its mechanism of action, as well as point out gaps in our current knowledge.

1. Discovery of kinetin and its activity in plants DNA (concentrations not reported) (Barciszewski et al., 1996). Later, K was also found in human urine (0.12 nM in urine of cancer patients, In 1955, Miller et al. isolated a new bioactive compound, N6-fur- 10–100 times less in healthy subjects) (Barciszewski et al., 2000), in furyladenine (Fig. 1A), from an autoclaved DNA sample from herring plant material (0.34 nM in coconut water) (Ge et al., 2005) and in brain sperm. Due to its ability to stimulate cytokinesis in various plant tissues, (up to 17.2 nM) and liver (up to 63.7 nM) tissue of transgenic mice the compound was given the name kinetin (K) (Miller et al., 1956, models of familial dysautonomia (Shetty et al., 2011). 1955a, b; Amasino, 2005). It was the first member to be identified from It should be pointed out that in a study conducted by our group, K the group of phytohormones which were later named cytokinins (Skoog was not detected in native extracts from Caenorhabditis elegans and et al., 1965). Chemically, cytokinins are adenines substituted with ei- Escherichia coli (Kadlecová et al., 2018). Orr et al. were unable to detect ther an isoprenoid or an aromatic side chain at the N6-position. Since K in native brain tissue of mice and rats (Orr et al., 2017). We have also their discovery, it has been established that cytokinins regulate plant been unable to identify K in various plant materials (unpublished data), growth, development, leaf senescence, apical dominance, seed germi- suggesting that the question of kinetin's natural origin is probably more nation, water and nutrient mobilization, flowering and many other complicated than was previously thought. processes (Kieber and Schaller, 2014). The proposed mechanism of formation of kinetin from DNA is based on a reaction of the amino group of adenine with the aldehyde group of 2. Occurrence of kinetin in nature furfural (Barciszewski et al., 1997a, b). The resulting base is then excised from the DNA backbone (Wyszko et al., 2003). The authors pre- For a long time, kinetin was believed to be an artificial product of sumed that furfural arises from oxidized deoxyribose by a mechanism DNA rearrangement (Hall and De Ropp, 1955), although some re- suggested by Pratviel and collaborators (Pratviel et al., 1991). Pratviel searchers suspected that it might be spontaneously formed in vivo from et al. did indeed detect furfural in DNA exposed to manganoporphyrin/

DNA (Skoog, 1994). The first evidence of its natural occurrence was KHSO5 artificial nuclease, but only when the samples were heated to reported in 1996, when K was identified in extracts from root nodules 90 °C for 15 min. However, as furfural is ubiquitously present in nature of Casuarina equisetifolia infected by the bacterium Frankia (Raman and and also occurs in various foods (Hoydonckx et al., 2000), it may be Elumalai, 1996). In the same year, Barciszewski et al. found K in plant present in cells from other sources. cell extracts (0.1 ng/g of dried material), in DNA isolated from human It has also been previously proposed that K can be formed in plants fibroblasts cultured in vitro and in commercially available calf thymus only under certain specific conditions, for example in wounded tissues

⁎ Corresponding author at: Laboratory of Growth Regulators, The Czech Academy of Sciences, Institute of Experimental Botany & Palacký University, Šlechtitelů 27, CZ-78371 Olomouc, Czech Republic. E-mail address: [email protected] (J. Voller). https://doi.org/10.1016/j.arr.2019.100958 Received 1 June 2019; Received in revised form 2 August 2019; Accepted 30 August 2019 1568-1637/ © 2019 Elsevier B.V. All rights reserved.

Please cite this article as: Alena Kadlecová, et al., Ageing Research Reviews, https://doi.org/10.1016/j.arr.2019.100958 A. Kadlecová, et al. Ageing Research Reviews xxx (xxxx) xxxx

Fig. 1. Structures of kinetin and kinetin ribotides. A) kinetin. B) kinetin riboside. C) kinetin riboside-5´-monophosphate. D) kinetin riboside-5´-triphosphate.

(Skoog, 1994) which undergo a spike in reactive oxygen species (ROS) keratinocytes, especially normal proliferation and differentiation is in- production. In humans, kinetin was detected at much higher con- dispensable for maintaining the epidermis and it is adversely affected centrations in the urine of cancer patients, suggesting that its formation by aging. Several studies reported that K treatment led to increased may be linked with the increased oxidative damage that accompanies expression of various differentiation markers in keratinocytes exposed the disease or its treatment (Barciszewski et al., 2000). If the formation to high levels of Ca2+ (Berge et al., 2006, 2008). The authors reported of K does, indeed, follow an increase in ROS production, this could that efficacy was greatest at concentrations of 40 and 80 μM(Berge partly explain its absence from native plant and animal/microorganism et al., 2006, 2008). Higher doses reportedly led to slower growth of samples in some studies. It might be fruitful to further investigate early passage cells, although the change was not accompanied by re- whether kinetin is present in extracts prepared from plant and animal duced viability or impaired DNA synthesis (Berge et al., 2006). In an- materials subjected to different types of stress, and compare them to other study, the effect of K on keratinocytes exposed to UVB irradiation native samples. Metabolization of kinetin excised from DNA or its ri- was investigated. The authors reported a 36% reduction in thymine bosides and ribotides by enzymatic deamination could also explain its dimer formation in the DNA of cells treated with 100 μMK (McDaniel absence from some materials. Cytokinins are substrates of cytokinin et al., 2005). oxidase/dehydrogenase in plants and adenosine deaminase in human Interestingly, K was also shown to modulate abnormal behavior in cells. primary keratinocyte cultures derived from psoriatic patients. It induced their differentiation, as demonstrated by reduced proliferation 3. Protective effects of kinetin in animal models rates, heterogeneous cell morphology and larger numbers of cells that developed cornified envelopes (Bolund et al., 1991). Clinical evidence When considering the effect of K in plants, researchers soon started that K might improve the symptoms of psoriasis is, however, still to wonder what activity it would have in other organisms. Initial ex- lacking. periments in tissue cultures derived from healthy human skin and in The beneficial effects of K have also been reported from 3D re- breast carcinoma connective tissue fibroblasts cultivated in vitro showed constructed skin equivalents (Vicanova et al., 2006). Here, topical ap- that K at low concentrations (approximately 0.05–0.3 μM) stimulated plication of 0.1% K resulted in induction of both proliferation (Ki-67) outgrowth but it had an inhibitory effect at higher concentrations and differentiation (filaggrin) markers. The authors only reported an (approximately 10–50 μM) (Orr and McSwain, 1957). The latter finding overall effect, without specifying in which keratinocyte layer pro- led to a further study of K's potential anti-cancer effect, but K had no liferation/differentiation occurred. In addition, K showed some activity effect on outgrowth of tissues derived from breast carcinomas (Orr and in deeper layers of the skin model. It stimulated the formation of the McSwain, 1960). The same, however, was not true for kinetin riboside basement membrane as demonstrated by increased levels of laminin 5. (KR, Fig. 1B) (Orr and McSwain, 1960; Hampton et al., 1956). Sub- The study also showed an increase in the content of elastin, fibrillin-1 sequent studies have also shown that some natural cytokinin ribosides and collagen type I in the elastic network in the upper dermis and that are highly cytotoxic to cancer cell lines while their corresponding bases they were organized perpendicular to the dermal-epidermal junction. have no or limited toxicity (Grace et al., 1967; Fleysher et al., 1969; Doležal et al., 2007; Voller et al., 2010, 2019). Nevertheless, cytokinin 3.1.1. In vivo studies and clinical trials bases including kinetin might still find a place in cancer therapy, as The reports of beneficial effects in skin cells led to an interest in the they induce differentiation in some leukemic cell lines (Ishii et al., application of K in cosmetic treatments and dermatology. During aging, 2002, 2003; Honma and Ishii, 2002). However, research on kinetin and skin undergoes structural and functional deterioration. This results in other cytokinin bases has focused primarily on their (cyto)protective increased risk of mechanical damage, exacerbated by a decrease in the activity, which will be discussed in more detail in the following sec- ability to induce wound healing responses, a higher risk of infection due tions. to decreased immune function and increased incidence of other skin disorders. The deterioration is enhanced by environmental factors, such 3.1. Anti-aging and protective effects of kinetin on skin as chronic UV exposure. Effects of K in vivo have been studied in 10-year-old dogs descended The anti-senescence activity of K in plants inspired Rattan and Clark from Mexican hairless dogs whose skin shows age-related changes si- to investigate its effect on aging in animal cells. Continuous cultivation milar to those occurring in humans. Dogs were treated topically, with of human mammary skin fibroblasts with K (40–200 μM) positively lotions containing from 10 to 10,000 μM and 2% K. Authors reported influenced several markers of aging, including cell size and mor- apparent improvements in skin texture, reduced wrinkling and a de- phology, levels of autofluorescence and cytoskeleton organization. K crease in pigmentation. After 100 days of treatment, improvements treatment helped late passage fibroblasts to successfully complete cy- were apparent even in the skin treated with the lowest K concentra- tokinesis, as indicated by a 10-fold reduction in the number of cells tions. Histological evaluation of the skin revealed a decrease in the containing multiple nuclei. No effect of K on growth rates and re- thickness of the corneal layer and a smaller number of melanin gran- plicative lifespan was reported in skin fibroblasts from two different ules. In dermis, the authors reported an increased number of collagen donors as well as in MRC-5 embryonic lung fibroblasts (Rattan and and elastic fibers and observed that their alignment was more orderly. Clark, 1994). K was also reported to increase the total amount of DNA Changes in pigmentation were reversible after discontinuation of the and [3H] uridine incorporation in human fibroblasts (Kowalska, 1992). treatment (Kimura and Doi, 2004). In a later study, the effect of a lotion In addition to skin fibroblasts, keratinocytes have also been used to containing 0.5% of K, as well as a commercially available cream (0.1% investigate the anti-aging effect of K. Proper functioning of of K), was evaluated on pig skin. The lotions were applied daily for 4

2 A. Kadlecová, et al. Ageing Research Reviews xxx (xxxx) xxxx days and treatment was followed by UV irradiation (at intensities of et al., 2017). Here, the authors re-introduced WT or mutant PINK1 into 5 mW/cm2 of UVB and 40 mW/cm2 of UVA). K did not reduce ery- the brain of PINK1 knock-out (KO) rats with the intention of studying threma generated in response to UV, nor did it reduce the number of the effect of K (approximately 300 mg/kg body weight/day, adminis- "sun-burned" cells in the skin. These results suggest that short-term tered in food). The model behaved rather surprisingly, as the authors application of K-containing cosmetic cannot substitute for the use of did not observe any loss of dopaminergic neurons in the original PINK1 sunscreen products (Tournas et al., 2006). KO animals, but they reported neuronal loss after reintroduction of both The safety and efficacy of K containing products has been assessed WT and mutant PINK1, suggesting unexpected toxicity on the part of in several clinical studies on human volunteers (McCullough and either PINK1 overexpression or the vector used. The authors also re- Weinstein, 2002; Wanitphakdeedecha et al., 2015; Chiu et al., 2007; ported a small, statistically insignificant trend towards an increase in McDaniel et al., 2005; An et al., 2017; Thornfeldt and Rizer, 2016). K, the number of dopaminergic neurons in K-treated animals and they although not necessarily the most active substance in comparative highlighted the need for development of a new in vivo model of this type studies, was generally reported to be well tolerated and to improve of familial PD. They also studied the effect of K in transgenic rats and parameters of photodamaged skin, including facial skin erythema, mice overexpressing α-synuclein. K (administered in food, 400–600 or moisture, hyperpigmentation and texture. Interestingly, a similar ben- 300 mg/kg/day on average for mice or rats respectively) did not show eficial effect and lack of adverse reaction could be observed not only in any protective activity in either model. In rats, the authors reported healthy subjects, but also in patients suffering from rosacea. The pa- similar rates of degeneration of dopaminergic neurons as in non-treated tients also reported significant improvement in several symptoms ac- control animals. K treated mice showed neither reduced levels of companying the disease, such as burning and stinging sensations and phosphorylated α-synuclein in hippocampus and cortex nor correction dryness of the skin (Wu et al., 2007). Some studies have also focused on of behavioral abnormalities. The lack of activity might be influenced by increasing the efficacy of K, either by combining it with other bioactive dosage, route of administration or induction of clearance mechanisms. compounds (Chiu et al., 2007; Maia Campos et al., 2012) or by in- The authors pointed out that the daily estimated dose rate for K decreasing its absorption by incorporating the molecule into solid lipid creased slightly over time as the animals gained weight, and they also nanoparticles (Goindi et al., 2015) or liposomes (Maia Campos et al., showed a significant 2 to 5-fold decrease in K levels in the brain tissue 2012). of the older animals (from the original 800–1000 pg/mg of tissue at the start of treatment) (Orr et al., 2017). It would be informative to further 3.2. Neuroprotective activity of kinetin investigate the effect of K in PD using different experimental settings and other models of both hereditary and sporadic PD. A decline in neuronal function and development of neurodegenerative diseases can undoubtedly be regarded as one of the most de- 3.2.2. Huntington´s disease (HD) bilitating signs of aging. The prevalence of neurodegenerative diseases A recent study also reported K as a potential modulator of HD is rising, mostly due to an increase in life expectancy, but the current (Bowie et al., 2018). Here, researchers showed that K is capable of in- treatment options are rather limited. K showed promising results in creasing the phosphorylation of the N17 subunit of mutant huntingtin studies focusing on the treatment of hereditary Parkinson’s disease in striatal cells derived from transgenic knock-in HD model mice. Such (PD), Huntington’s disease (HD), familial dysautonomia (FD) and neu- modification was shown to reduce the detrimental changes accom- rofibromatosis 1 (NF1), as well as in models utilizing stressors for in- panying the disease in both in vitro and animal models of HD (Gu et al., duction of certain symptoms of (neuro)degeneration. 2009; Atwal et al., 2011). As discussed below, the mechanism of action involves an increase in the activity of casein kinase 2. In further ex- 3.2.1. PINK-1 activation and Parkinson’s disease (PD) periments, 0.1–1 μM K increased the viability of mouse cortical neurons K showed a promising effect in in vitro models of the hereditary form transfected with a fragment of mutant huntingtin in a dose dependent of early onset PD, caused by a loss-of-function mutation (G309D) in manner. Next, the authors studied the effect of K in transgenic HD PTEN-induced kinase 1 (PINK1) (Hertz et al., 2013). This kinase plays a model mice expressing mutant human huntingtin containing 128 CAG crucial role in mitochondrial quality control by targeting depolarized repeats. The characteristic phenotype of this model is impairment of mitochondria for degradation by a Parkin-dependent mechanism. The motor functions and anxiety-like behavior. Animals receiving 0.83 or authors showed that a metabolite of K, kinetin riboside-5´-triphosphate 4.17 mg/kg/day of K intraperitoneally (i.p.) fared better in tests mea- (KRTP), is able to function as a neo-substrate of PINK-1, boosting its suring their motor performance on a rotarod. The effect was dose de- activity (details below). The authors pre-treated human neuroblastoma pendent. To assess the level of anxiety, the authors chose to administer cell line SH-SY5Y with 50 μM K and then exposed the cells to carbonyl K in the diet (653 mg/kg of chow; concentrations of K in animal tissues cyanide m-chlorophenyl hydrazone (CCCP), a toxin which causes de- were not reported) to avoid stress caused by regular painful injections. polarization of the mitochondrial membrane. In K-treated cells, they K prevented anxious behavior in 2 out of 3 tests carried out. In mice fed observed a PINK1-dependent increase in phosphorylation of Bcl-xL. In with K, the authors did not observe any improvement in motor func- HeLa cells transfected with either wild-type (WT) or mutant PINK1, tion, probably due to the lower dose and decreased absorption of K treatment with K resulted in faster PINK1-dependent recruitment of using this route of administration. Orally treated animals showed a Parkin, and in an increase in its phosphorylation levels. In SH-SY5Y and decrease in levels of insoluble mutant huntingtin in the cortex, but not Parkin-transfected HeLa cells pre-treated with K, the authors reported in the striatum. A similar trend was observed in mice receiving i.p. reduced cleavage of caspase 3/7 following exposure to a proteasome injections, with a small, statistically insignificant decrease in huntingtin inhibitor. This is in agreement with yet another described biological levels in the striatum as well (Bowie et al., 2018). activity of PINK1 – its anti-apoptotic effect in cells undergoing proteasomal stress (Klinkenberg et al., 2010). A smaller number of apop- 3.2.3. Familial dysautonomia totic SH-SY5Y cells was also detected. K also decreased the mobility and K also showed a protective effect in yet another hereditary neuro- velocity of axonal mitochondria in rat hippocampal neurons – a de- degenerative disorder – familial dysautonomia (FD). This rare but decrease in these parameters is considered to be a first step in PINK1- bilitating disease is caused in more than 99% of cases by a point mu- regulated removal of damaged mitochondria. Furthermore, a K con- tation (T → C/ IVS20 + 6T→C) in the splice site of intron 20 of the centration of 50 μM had no demonstrable toxicity on cultured dopa- gene encoding elongator complex protein 1 (ELP1). This leads to de- minergic neurons even after a relatively long period of treatment (10 creased splicing efficiency and skipping of exon 20 in a pre-mRNA days). transcript, resulting in synthesis of truncated ELP1. The protein reg- A subsequent study attempted to validate these results in vivo (Orr ulates transcription of multiple genes, including some encoding

3 A. Kadlecová, et al. Ageing Research Reviews xxx (xxxx) xxxx proteins involved in cytoskeleton function and cell movement which clinical studies so far are encouraging, especially given that it contains a are necessary for normal cell growth and development. The mutation furan moiety. The furan moiety is a potential toxicophore as it can be does not have full penetrance and while some tissues are able to com- metabolized by cytochrome P450 enzymes to produce reactive elec- pensate and produce a sufficient amount of functional protein, nervous trophilic species (Bakhiya and Appel, 2010). Many furan-containing tissue is severely affected. FD is therefore characterized by impaired compounds are therefore hepatotoxic (Peterson, 2012). development and survival of neurons in both sensory and autonomic The ability of K to modulate aberrant splicing, although highly nervous system, resulting in progressive neurodegeneration (Shohat specific as discussed below, is not limited solely to familial dysauto- and Hubshman, 2014). nomia. Another disease frequently linked with mutations leading to In 2004, Slaugenhaupt et al. found that K could correct the aberrant aberrant splicing (in 30–40 % of cases) is neurofibromatosis type I splicing of ELP1 in a lymphoblast cell line derived from FD patients. (NF1). This disorder is characterized by the presence of "café au lait" The effect was dose dependent, with the highest non-cytotoxic con- spots and neurofibromas. Less common but more serious manifestations centration tested (400 μM) being the most effective. The results were include, for example, the development of malignant tumors (Friedman, consistent across various cell lines (Slaugenhaupt et al., 2004). A si- 2018). In their study, Hims et al. created an NF1 model by transfecting milar effect was observed in a subsequent study, where the authors human embryonic kidney (HEK) cells with minigenes containing part of demonstrated that treatment of FD lymphoblast cells with 100 μMK the WT or mutated NF1 sequence. One of the mutant constructs mod- restored the amount of WT ELP1 at both mRNA and protein level to that elled the splicing defect and was responsive to K – after treatment, the similar to non-treated controls. Interestingly, K also increased the levels authors observed increased incorporation of exon 36 (Hims et al., of WT ELP1 mRNA in cell lines derived from carriers of the disease and 2007). In a later study, Pros et al. examined the effect of K in both short from healthy controls (Hims et al., 2007). Although a later study re- term and immortalized lymphocyte cell lines derived from 22 patients ported an increase in WT ELP1 only at the mRNA and not at the protein carrying 19 different NF1-splicing defects. A decrease in aberrantly level, it is possible that the relatively short 3-day treatment may have spliced product following treatment with 100 μM K was observed for 4 accounted for the difference (Sinha et al., 2018). different mutations. The effect was also reproducible in patient-derived A positive effect of K on ELP1 expression was also observed in cells fibroblasts, suggesting that it is not limited to a specific cell type (Pros of neuronal lineage, including olfactory ecto-mesenchymal stem cells et al., 2010). It must be noted, however, that unlike in FD, where the collected from FD patients (Boone et al., 2010), neural crest precursors vast majority of patients harbor the same mutation, more than 1000 (Lee et al., 2009) and sensory neurons derived from induced pluripotent different mutations that cause NF1 exist and K treatment would stem cells created from fibroblasts of FD patients (Zeltner et al., 2016). therefore be of potential benefit to only a small proportion of patients. Treatment also increased the differentiation of neural precursors (Lee et al., 2009) and protected sensory neurons against degeneration 3.2.4. Protective activity of kinetin in models of CNS stress (Zeltner et al., 2016). The effective concentrations ranged from 25 to A common characteristic of the majority of neurodegenerative dis- 200 μM. On the other hand, K did not correct the migration deficits of eases is an increase in oxidative stress (Liguori et al., 2018). K has also the cells (Lee et al., 2009; Boone et al., 2010). been reported to protect neurons and nervous tissue against various Correction of aberrant splicing by K was also observed in vivo.In type of stressors. mice expressing humanized FD and WT ELP1 and fed with 400 mg/kg/ Radhakrishna et al. investigated the effect of K on radiation-induced day of K for 30 days, Shetty et al. reported an increase in exon 20 in- behavioral changes (Radhakrishna et al., 2017). Mice, orally treated clusion in various tissues, including brain, liver, kidney, blood, spleen, with K for 5 days, showed increased survival following exposure to a lung, heart and eye. K was detected in the serum, liver and brain tissue lethal dose of radiation (10 Gy), with 100 mg/kg of K being the most (Shetty et al., 2011). This suggests that K is absorbed and, importantly, effective dose. This dose was then used in subsequent experiments, in can even cross the blood-brain barrier. which they evaluated behavioral deficits in mice exposed to lower, sub- The promising results of these studies, as well as favorable results in lethal doses of radiation (6 Gy). The treated animals displayed reduced pre-clinical pharmacokinetics and safety studies in rodents (Axelrod anxiety levels and improved learning ability. The authors speculated et al., 2011), eventually led to clinical trials. The pharmacokinetics, that this could be due to the protective effect of K in dopaminergic tolerability and efficacy of oral K treatment were first evaluated in 29 neurons. asymptomatic heterozygous carriers of the mutation. The treatment The effect of K was also investigated in glial cells, specifically in a lasted for 8 days and K was supplied once per day at doses ranging from primary culture of rat astrocytes. Treatment with K (50–200 μM) pro- 3.35 to 23.5 mg/kg/day. The majority of the subjects receiving tected the cells against D-galactose (D-gal) induced toxicity and pre- 23.5 mg/kg/day of K reached the desired plasma levels of 2150 ng/mL vented their morphological deterioration (Liu et al., 2011). Exposure to (10 μM). K was quickly absorbed and eliminated according to first order high D-gal doses leads to glycoxidative damage and degenerative elimination kinetics. The treatment led to a dose dependent increase in changes resembling those occurring with aging (Sadigh-Eteghad et al., normal ELP1 mRNA in peripheral leukocytes. The most common ad- 2017). verse effect was mild to moderate nausea. More serious adverse effects The (neuro)protective effect of K was also studied in mice receiving included severe nausea, vomiting, diarrhea, rash or headache and re- a treatment in which D-gal (subcutaneous injections, 100 mg/kg body sulted in 4 subjects withdrawing from the study after administration of weight) was combined with AlCl3 (0.1088 mg/mL in drinking water) the first dose (Gold-von Simson et al., 2009). (Wei et al., 2017). Some researchers use aluminum to induce neuro- In a follow-up study, 8 FD patients received 23.5 mg/kg/day of K for degeneration with some features resembling those of Alzheimer´s dis- 28 days in total. In the 7 patients who completed the study – one of ease, although the relevance of such a model is questionable (Lidsky, them withdrew due to reasons unrelated to the treatment – the mean 2014). In this study, 5, 10 and 20 mg/kg of K were orally administered improvement in exon 20 inclusion was 17% in white blood cells. All to the mice. After 90 days of treatment, researchers evaluated the patients except one reached the target plasma levels. Side effects were spatial learning and memory of mice and showed improvement in K mild, and included two cases of diarrhea, three cases of mild nausea and treated groups. Further experiments showed that K attenuated several three cases of headache. Elevated liver enzymes in three patients and histopathological changes in the CA3 region of the hippocampus, which one case of decrease in platelet and white blood cell count were also plays a crucial role in spatial memory, restored acetylcholine content reported. These changes were reversed after treatment was dis- and decreased the activity of acetylcholinesterase. The authors also continued. According to ClinicalTrials.gov, K is currently in phase II of showed reduced levels of Aβ 1–42 containing plaques in the cortex and clinical trials (Identifier: NCT02274051). hippocampus and a decrease in levels of expression of amyloid pre- Kinetin´s favorable toxicity profiles observed in preclinical and cursor protein, β-secretase, γ-secretase and amyloid beta 1–42. The

4 A. Kadlecová, et al. Ageing Research Reviews xxx (xxxx) xxxx effect was dose dependent in all experiments. apoptotic protein Bcl-2 to the anti-apoptotic protein Bax was similar to In a recent study, Wei et al. used the immortalized mouse hippo- that in healthy control rats (Li et al., 2014a,b,c). K also protected the campal cell line HT22 to assess the ability of K to prevent glutamate- reproductive organs of D-gal treated female mice (Sun et al., 2013). induced oxidative damage. The authors observed that 8 -h pre-incuba- Animals receiving 25 or 50 mg/kg/day of K by intragastric injections tion of the cells with 5 mg/mL (approximately 23 μM) of K protected for 40 days showed a decrease in atrophy of the ovaries and uterus. them against toxicity induced by 5 mM of glutamate, as demonstrated Mice also displayed higher levels of estrogen, a shorter estrous cycle by increased cell viability, a decrease in the number of apoptotic and and an increased number of mature oocytes. necrotic cells and reduced cytolysis. They also reported inhibition of apoptosis signal-regulating kinase 1 and other downstream members of 3.4. Lifespan studies in invertebrate models its signaling pathway. Moreover, K-treated cells maintained normal mitochondrial function – the treatment prevented mitochondrial Reduced rates of aging following K treatment are not limited to cell membrane depolarization and restored the ATP content (Wei et al., cultures and rodent models of accelerated aging; they have also been 2018). reported in invertebrate models. Fruit flies of the species Zaprionus paravittiger and Zaprionus indianus fed with 25 ppm K (approximately 3.3. Protective effect of kinetin in cardiovascular and other systems 120 μM) showed prolonged lifespan, slower development (Sharma et al., 1995, 1997) and decreased fertility (Sharma et al., 1997). Higher Cardiovascular diseases are a leading cause of death worldwide. concentrations either resulted in smaller beneficial effects or were toxic Here too, aging is a major risk factor. Age-related deterioration of en- to the animals (Sharma et al., 1995). 200 μM K significantly increased dothelial cells contributes to a decrease in the plasticity of blood vessels longevity in Caenorhabditis elegans (Kadlecová et al., 2018). The K- and is linked to the development of vascular diseases (Donato et al., treated worms also showed greater resistance to oxidative and heat 2015). Interestingly, continuous cultivation with 50 μM K was reported stress. to have a protective effect in dermal microvascular endothelial cells It is possible that the effect of kinetin on aging in different models HDMEC. K protected the cells against apoptosis and increased the may be due to the induction of the same, highly conserved mechanism. number of dividing cells in the culture. The late passage cells main- However, we currently lack definitive experimental evidence of the tained morphology resembling those in the earlier passages. There were exact mechanism, apart from direct radical quenching which, as dis- also fewer senescent cells. As a result, 7 more population doublings cussed below, might be responsible. As K is probably a multi-target were achieved (Lee et al., 2006). molecule, it is possible that its effects in various models, ranging from In another study, K inhibited platelet aggregation in suspensions dividing human cells to post-mitotic worms, could result from different activated by several different inducers in a dose-dependent manner mechanisms. (Sheu et al., 2003). A subsequent investigation demonstrated that a similar effect can also be observed in vivo. Intravenous application of 4 4. Mechanism of action or 6 mg/kg of K increased the number of platelets in plasma and prevented death in mice with acute pulmonary thrombosis induced by As summarized in the previous sections, various protective effects of ADP. The same dosage of K also increased bleeding time in rats with K in diverse models have been reported. However, our knowledge about severed mesenteric arteries. Moreover, 13–16 mg/kg of K prolonged the the mechanism of action remains relatively limited. It might be time needed for thrombus formation in microvessels of mice exposed to tempting to create hypothetical links between the activity of K in ani- fluorescein and light irradiation (Hsiao et al., 2003). mals and that in plants, especially as it seems that some of its effects – Some evidence that K might positively influence the symptoms of such as the promotion of differentiation and protection against oxida- certain metabolic diseases, such as diabetes, also exists. Treatment with tive stress – are shared between both kingdoms. But such reasoning is K resulted in an increase in glucose uptake in cultured myocytes supported by little experimental evidence. Apart from the other vast (0.3–7.5 μM) and in ex vivo epitrochlearis rat muscle (0.5 and 2.5 μM). molecular and physiological differences, the receptors and signaling The authors of this patent also prepared cytokinin-enriched extracts systems that mediate the activity of cytokinins in plants are not present from sprouted barley. The extracts had a beneficial effect on glucose in animals (Voller et al., 2017). absorption in vitro, and, remarkably, decreased blood glucose levels in rats in which diabetes was induced by streptozocin. The same extracts 4.1. Protection against oxidative stress were then orally administered to patients with type 2 diabetes for 90 days, resulting in decreased serum glucose levels, increased glucose In many studies, the effect of K has been ascribed to its ability to tolerance, reduced levels of glycosylated hemoglobin and an improved reduce oxidative stress. Reactive oxygen species (ROS) play an im- LDL/HDL ratio. Although the authors note that they confirmed the portant role in the pathogenesis of multiple diseases and possibly also in presence of K in the extract by HPLC and LC/MS analysis, they did not aging (Liguori et al., 2018). They have the ability to damage macro- report the exact concentrations of various cytokinins and they mea- molecules, which, over time, might lead to a deterioration at the cel- sured only the beneficial e ffect of the combination of all phytochem- lular, and subsequently tissue and organ, levels (the free radical theory icals present in the extracts (Mijikovic et al., 2007). The extent to which of aging; see Harman, 1955; Perez et al., 2009)). Although we now kinetin contributed to the observed effects is therefore unclear. know that ROS also act as essential signaling molecules (Davalli et al., Other studies were conducted with D-gal treated rodent models of 2016), the fact remains that any decline in cellular protection me- accelerated aging, focusing on organs other than the brain. In rats ex- chanisms, resulting in excessive amounts of ROS, is detrimental. posed to D-gal, administration of K in subcutaneous injections for 45 Initially, it was proposed that K acts as an antioxidant. It could di- days delayed atrophy of the spleen, with the highest dose used, 20 mg/ rectly scavenge ROS in several possible ways. It might form complexes kg/day, being the most effective. The histological profile and appear- with copper that have superoxide dismutase-like activity (Parvez and ance of the spleen cells were more similar to those of healthy control Birdsall, 1990). The radicals could also abstract hydrogen from the α- animals. K also restored serum levels of interleukin-2 and interleukin-6, carbon of the amine bond (Barciszewski et al., 1999) or possibly from which play pivotal roles in immune response modulation. Increased the electron-rich furan ring. Furan derivatives are known scavengers of levels of several immunoglobulins (IgA, IgM, IgG) were detected. In ROS (Okada and Okajima, 1998; Lemke et al., 2014; Okada et al., spleen lymphocyte suspension isolated from the animals, the authors 1996). The extent of kinetin's ROS scavenging capability has been detected fewer apoptotic and more proliferating cells. Mitochondria measured in multiple studies using several different biochemical assays. maintained their membrane potential, and the ratio of the pro- Brizzolari et al. reported that K had some radical scavenging activity in

5 A. Kadlecová, et al. Ageing Research Reviews xxx (xxxx) xxxx an ORAC (oxygen radical absorbance capacity) assay and a deoxyribose Lipids undergo oxidative damage as well. Markers of lipoperoxida- degradation assay but did not observe any eﬀect in a TEAC (Trolox tion include lipid hydroperoxides (LP) and malondialdehyde (MDA). In equivalence antioxidant capacity) assay (Brizzolari et al., 2016). In vitro, treatment with 100 μM K completely prevented the formation of another study, the antioxidant capacity of K was measured using a LP in low density lipoproteins isolated from human blood and induced

Photochem system. Here, K scavenged ROS only at a concentration of by CuSO4. It also slightly reduced MDA formation (to 76% of the con- 1 μM, the highest one tested, and performed less well compared to trol) in microsomal preparations from rat liver induced by the pro- several other known antioxidants such as idebenone, DL-α-tocopherol, oxidant system NADPH/ADP/Fe3+(McDaniel et al., 2005). Some de- L-ascorbic acid or ubiquinone. No intrinsic antioxidant activity of K was crease in levels of membrane lipoperoxidation products was observed in observed in a FRAP (ferric reducing antioxidant power) assay at con- K treated fibroblasts (Jabłonska-Trypuc et al., 2016). In this case, the centrations of up to 0.5 μM(Othman et al., 2016). In a recent study, the effect was observed in vivo as well – decreased levels of lipoperoxidation ROS scavenging ability of K was measured in 5 different assays – DPHH markers were detected in the brain (Wei et al., 2017; Liu et al., 2011), (1,1-diphenyl-2-picrylhydrazyl), FRAP, Fe2+ chelation, superoxide ra- spleen (Li et al., 2014a,b,c) and reproductive organs of mice exposed to dical inhibition and hydroxyl radical inhibition assays. K performed D-gal (Sun et al., 2013). poorly in all of them (Wei et al., 2018). Taken together, these results In a study focused on late-passage endothelial cells after long-term suggest that although K probably has some ROS scavenging effect, it is cultivation with K, Lee et al. described some K-mediated changes that unlikely to be a particularly strong antioxidant. took place at the protein level. Utilizing 2-D electrophoresis, they Apart from scavenging ROS directly, K could also protect cells by showed several dozen protein spots altered in K-treated cells and then inducing anti-oxidative enzymes, such as superoxide dismutase (SOD), identified some of these proteins by MALDI-TOF MS. Differentially catalase (CTL) or glutathione peroxidase (GP). Protection by low doses expressed proteins included those involved in antioxidant defenses, of K (100 nM) against glutathione depletion caused by the stressor cytoskeleton function, intracellular trafficking, cell-cycle progression, patulin, as well as reduced levels of intracellular ROS following 4-ni- translation and protein turnover, coagulation and collagen maintenance troquinoline 1-oxide treatment, were observed in the human promye- (Lee et al., 2006). locytic cell line HL-60 (Othman et al., 2016). K also suppressed ROS production in collagen-activated platelets (Hsiao et al., 2003). In the 4.2. Hormetic activity mouse hippocampal cell line HT22, K decreased the ROS levels as well. This may be ascribed to the reported translocation of the transcription It has also been proposed that kinetin can act as a hormetin (Rattan, factor Nuclear factor (erythroid-derived 2)-like 2 (Nrf2) to the nucleus 2002, 2008). Hormesis can be defined as a beneficial biological effect and subsequent expression of its targets, including heme oxygenase-1 which appears after stimulation of cellular protective mechanisms by (HO-1) (Wei et al., 2018). Nrf2 is a major regulator of xenobiotic me- mild stress (Mattson, 2008). As outlined above, K has been reported to tabolism, oxidative and electrophilic stress responses, etc., and its target stimulate antioxidant enzymes in multiple models, and it induced heat- genes include a number of protective enzymes, such as HO-1 (Ma, shock proteins (Hsp27 and Hsp70) and HO-1 in human keratinocytes 2013). (Berge et al., 2008). Activation of the Nrf2 pathway in mouse hippo- Interestingly, the content of HO-1, as well as the activity of SOD, campal cells (Wei et al., 2018) could be a result of kinetin´s hormetic CTL and GP, was also increased in the brains of K and D-gal treated effect. How exactly kinetin stimulates the hormetic pathways is not mice (Wei et al., 2017). These results are in agreement with those of an clear at this point. The only insight into the mechanism of action comes earlier study, in which the authors also reported induction of SOD and from a study on C. elegans. We observed that co-treatment of these GP in brain tissue of D-gal and K treated animals (Liu et al., 2011). worms with 200 μM K and 100 μM of the antioxidant Trolox led to Increased activity of antioxidant enzymes was also observed in the re- suppression of kinetin's ability to prolong the lifespan and induce stress productive organs (Sun et al., 2013) and spleen of rodents exposed to K resistance (Kadlecová et al., 2018). This suggests that the presence of and D-gal (Li et al., 2014a,b,c). ROS is necessary for kinetin to have this effect. As outlined above, increased oxidative stress is associated with An interesting question is whether it is K itself or some of its me- damage to macromolecules. K has been reported to protect DNA against tabolites that are responsible for the hormetic activity. In worms, we Fenton reaction-mediated oxidative damage in a biochemical assay. observed efficient metabolization of K into kinetin riboside (KR) and Kinetin (100 μM) significantly reduced the formation of 8-hydroxy-2- kinetin riboside-5´-monophosphate (KRMP) (Kadlecová et al., 2018). deoxyguanosine, which results from hydroxyl radical attack on the C8 Kinetin ribotides were also detected in human cell lines after treatment of guanine and is frequently used as a marker of oxidative DNA damage. with K (Hertz et al., 2013; Bowie et al., 2018; our unpublished data). KR The authors ascribed this reduction to either the intrinsic antioxidant is a known cytotoxic compound which induces ATP depletion in cells activity of K, or its possible interaction with iron, which is used as a (Cabello et al., 2009), possibly by impairing mitochondrial function catalyst in the reaction (Olsen et al., 1999). Pre-treatment with 100 nM (Cheong et al., 2009). Formation of a small amount of riboside in cells K reduced the number of DNA breaks after exposure to the genotoxic after K treatment could therefore cause mild oxidative stress and trigger agent 4-nitroquinoline 1-oxide in freshly isolated human leukocytes and an adaptive response. The fact that other cytokinin ribosides in low 3different cell lines: HL-60 (a human promyelocytic cell line), HaCat (a doses have been previously reported to act as Nrf2 inducers (Dassano human keratinocyte cell line) and NRK (an epithelial rat kidney cell et al., 2014) provides some support for this theory. We also observed line) (Othman et al., 2016). Reduction in levels of 8-hydroxy-2-deox- induction of HO-1 after KR treatment of skin fibroblasts and keratino- yguanosine was also observed in vivo, in the brains of K treated mice cytes (unpublished). It is necessary to note that low doses of other exposed to D-gal (Wei et al., 2017). metabolites – such as those formed by metabolization of the furan ring K has also been shown to prevent protein oxidation and glyoxida- present in the molecule – could possibly act as hormetins. Electrophilic tion in vitro. Verbeke et al. incubated bovine serum albumin with glu- α,β-unsaturated dialdehydes known to arise from furan ring opening cose – a less reactive sugar causing slower glycation/glyoxidation – or a (Peterson, 2012) may also activate Nrf2. more aggressive combination of ribose, arabinose and glyoxal. K (50 Explaining any hormetic mechanism of action has its possible pit- and 200 μM) prevented the formation of advanced glycation age pro- falls. The hormetic concentration – i.e.stress inducing but not yet toxic – ducts (AGEs) and carbonylated proteins. K also prevented aggregation may differ not only between individual cell and tissue types but also of proteins, which results from the cross-linking and fragmentation between the same cells/tissues under various physiological conditions. induced by the damage (Verbeke et al., 2000). Remarkably, formation If kinetin is indeed a hormetin precursor (a pro-hormetin), differences of AGEs in D-gal treated mouse brain tissue was also prevented by K in metabolism among various cells/tissues could further complicate the treatment (Wei et al., 2017). picture. Cells/tissues may also differ in their ability to accumulate the

6 A. Kadlecová, et al. Ageing Research Reviews xxx (xxxx) xxxx

(pro-hormetin. Differences in the accumulation of kinetin in mouse did not observe any protection against various mitochondrial toxins, brain and liver tissue have already been reported (Shetty et al., 2011). including CCCP, conferred by kinetin (unpublished data). However, analysis of a broader spectrum of tissue types, including de- The KRTP content can be increased by using prodrugs aimed at tailed analysis of kinetin metabolites, is still lacking. delivering KRMP to cells, thus overcoming the limited efficiency of the base transformation. A recent study described the development of 4.3. Kinetin as a mediator of stress response ProTides (one of the prodrug types developed for delivery of nucleoside 5′-monophoshates into cells) designed to deliver KRMP into cells, and As K is widely considered to be a natural molecule, formed in DNA their ability to activate PINK1 in HEK cells (Osgerby et al., 2017). The during oxidative damage, it was previously proposed that its formation authors reported that KR at a concentration of 50 μM can also activate could itself be a protective response to a stress stimulus (Barciszewski PINK1. This is not unexpected, since it was previously reported that et al., 1997b; Maiuri et al., 2018). The presence of K in the DNA could cells are able to efficiently transform KR into KRMP (Mlejnek and possibly activate DNA repair enzymes. K has also been proposed to act Doležel, 2005), but high cytotoxicity and low bio-availability make KR as an antioxidant, responding to the oxidative stress that triggered its unsuitable as a drug candidate. formation in a negative feedback loop. However, as discussed above, K In their patent, Mijikovic et al. showed that another kinase, AMP- was reported to perform rather poorly in multiple in vitro assays mea- activated protein kinase (AMPK), was regulated by kinetin (Mijikovic suring its ROS scavenging ability. Moreover, K was typically detected at et al., 2007). This enzyme detects intracellular AMP/ADP/ATP ratios nM concentrations in native tissues (Barciszewski et al., 2000; Shetty and subsequently regulates a number of metabolic processes. It also et al., 2011). Tens or even hundreds of μM of K are usually required for plays an important role in the modulation of aging and dysfunction of a protective effect in in vitro assays and doses of several mg/kg/day the enzyme is implicated in multiple diseases (Jeon, 2016). Using im- need to be administered in in vivo experiments. Even when considering munoblotting, the authors reported that treatment of cultured myocytes the possible increase in formation of K under stress and the disturbance with 0.1–10 μM K led to an increase in the phosphorylation of Thr172 of the intracellular environment that accompanies pathological states, it (i.e.activation) of AMPK. Activation of two of its downstream targets, should be carefully considered whether naturally formed K (and the protein kinase B and glucose transporter 4, was also observed. Inter- KRTP that may form subsequently) could reach sufficient levels to in- estingly, KR was identified as an AMPK activator as well. It is tempting duce a meaningful protective response. to speculate that metabolization of K and KR to KRMP is required for APMK activation. It was previously suggested that another cytokinin, 4.4. Interaction with protein kinases N6-isopentenyladenosine-5´-monophosphate, is able to bind to the gamma subunit of AMPK, functioning as its activator instead of AMP Hertz et al. proposed a very exciting mechanism of action for K in (Pisanti et al., 2014). It is therefore possible that KRMP might have a their study focused on hereditary PD caused by mutations in PINK1. similar effect. It should be noted, however, that activation of AMPK They suggested that the active compound is kinetin riboside-5´-tri- may also be due to a decrease in ATP levels caused by cytokinin ribo- phosphate (KRTP, Fig. 1D), which acts as a PINK1 neo-substrate. In in sides. vitro assays, they showed that instead of ATP, KRTP can act as a substrate for PINK1, boosting its activity (Hertz et al., 2013). A later study 4.5. Splicing modulation based on molecular modelling showed that this might be due to the N6- furfuryl group causing steric hindrance to the main chain of the hinge The exact mechanism by which K modulates splicing in familial region connecting the N- and C-lobes. This might influence their spatial dysautonomia and neurofibromatosis 1 is not yet fully understood. It arrangement and subsequently the kinase activity (Okatsu et al., 2018). has been shown that K does not generally increase inclusion of alter- Because charged KRTP cannot cross cell membranes, Hertz et al. used K natively spliced exons (Slaugenhaupt et al., 2004; Hims et al., 2007). in cell-based assays and demonstrated that it can be transformed first Moreover, microarray evaluation of the activity of K in olfactory ecto- into kinetin riboside-5´-monophosphate (KRMP, Fig. 1C) by adenine mesenchymal stem cells showed that K altered the expression of only a phospho-ribosyl transferase, and then to higher phosphates, evidenced rather small number of genes. Among these were SNRPA, encoding a by the fact they detected KRTP in HeLa cells. Moreover, K boosted the core component of the snRNP U1, and LUC7L, which encodes a putative activity of wild-type PINK1. It is an interesting question whether this subunit of snRNP U1. The authors suggested that K reinforces 5’ splice activation of wild-type PINK1 could protect cells against oxidative site (ss) recognition by improving recruitment of splicing factors (Boone stress due to more efficient maintenance of a healthy mitochondrial et al., 2012). In their study aimed at finding a sequence within exon 20 population, or whether it could prove detrimental in the end. Never- required for sensitivity to kinetin, Hims et al. created a library of exon theless, the work of Hertz et al. suggests an interesting new strategy – 20 deletion constructs. They pinpointed the critical sequence as being using neo-substrates to increase the activity of kinases – which could the last 3 nucleotides, a CAA motif, on the 5’ ss itself. The authors also potentially be utilized in treating many diseases (Hertz et al., 2013; examined whether splicing of other genes could be modulated by K. Kleiner and Kapoor, 2013). A similar idea was adopted in a study by They selected over 40 candidates based on a literature search for Maiuri et al. focused on Huntington´s disease. They proposed that KRTP transcripts with a single alternatively spliced exon and on unpublished serves as a more efficient neo-substrate for casein kinase 2, an increase microarray data, where they identified genes differentially expressed in the activity of which leads to phosphorylation of the N17 subunit of after K treatment. They found two genes – BMP2 and ABI2 – that mutant huntingtin, and this has a protective effect (Maiuri et al., 2018). showed small but consistent increases in internal exon inclusion fol- Because KRTP is consumed during the kinase reaction, to have any lowing treatment with K. In both cases the genes possess the CAA motif meaningful protective activity in the cells its initial concentration must on 5’ ss. The K-responsive NF1 construct also shared the same motif, be sufficiently high in order to compete with ATP for the binding site of although misplaced by one base (Hims et al., 2007). A later study the kinase. Although Hertz at al. demonstrated that KRTP can effec- identified one more sensitive transcript, ZNF280D, which has the same tively compete with ATP as a PINK1 substrate in vitro if it is present at a 5’ ss motif (Boone et al., 2012). concentration 4 times lower than that of ATP, in HeLa cells they de- It is somewhat surprising that the microarrays showed only a lim- tected it at a 200-times lower concentration (68 ± 13 μM, compared to ited number of differentially expressed genes (Boone et al., 2012), given 1950 ± 421 μM of ATP) (Hertz et al., 2013). In our own experiments, that due to its structure, K could be expected to interact with various using capillary electrophoresis we detected even lower concentrations members of the cell´s purinome and subsequently modulate the activity of KRTP (< 10 μM) in 5 different cell lines treated with 100 μM K for of multiple pathways. We expect that Next-Generation Sequencing ex- 24 h. Low intracellular concentrations of KRTP could explain why we periments will better capture transcriptome changes induced by

7 A. Kadlecová, et al. Ageing Research Reviews xxx (xxxx) xxxx

was shown to induce melanogenesis by activation of protein kinase A and subsequent up-regulation of microphthalmia-associated transcription factor and tyrosinase expression (Kim et al., 2009). We observed a similar effect for para-topolin (unpublished data). Some derivatives of K have been developed and their protective effects investigated in animal models as well. A notable example is Pyratine® (6-furfurylamino-9-(tetrahydropyran-2-yl)-9H-purine; a 9- tetrahydropyranyl derivative of kinetin, Fig. 2E), another commercially used cosmeceutical, which reduced the symptoms of rosacea and improved the appearance of photodamaged skin in clinical trials (McCullough et al., 2008; Tremaine et al., 2010; Ortiz et al., 2009). Recently, our institute developed more efficient cytokinin derivatives able to protect cells against UV irradiation in vitro and increase the stress resistance of worms (Hönig et al., 2018). Our laboratory is involved in several projects focused on the development of cytokinin- inspired molecules with protective activity, with possible applications in treating skin and metabolic disorders. Our aim is to prepare more active derivatives, as well as to improve their pharmacokinetics Our recent results show that various cytokinin derivatives either protect against, or sensitize C. elegans to, oxidative and/or heat stress. The protective activities of cytokinin derivatives with substitutions at position C8 of the purine ring seem to be especially promising (Fig. 3). Several of these active compounds also protected fibroblasts derived from patients with Friedreich ataxia against oxidative stress (unpublished, Fig. 3). Friedreich ataxia was selected as a model of mitochondrial disease, because of its complex phenotype including oxidative stress, ATP depletion and mitochondrial iron overload. However, Fig. 2. Structures of natural and synthetic cytokinins. A) trans-zeatin. B) N6- we envision the use of cytokinins in treating not only Friedreich ataxia isopentenyladenine. C) N6-benzylaminopurine. D) para-topolin. E) 6-furfur- but also other mitochondrial diseases. ylamino-9-(tetrahydropyran-2-yl)-9H-purine (Pyratine) F) 2-chloro-N-(furan-2- The ability of K to correct aberrant splicing in FD models and pa- ylmethyl)-7H-purin-6-amine (RECTAS). tients has led to increased interest in the development of more active analogs. The first reported compound showing approximately 25-fold kinetin, including splicing modulation. Such studies could also provide higher efficacy than K was 2-chloro-N-(furan-2-ylmethyl)-7H-purin-6- insights into whether kinetin's protective effect and ability to modulate amine, also known as RECTAS (Fig. 2F) (Yoshida et al., 2015). In a aging could be related to alternative splicing of pre-mRNA. recent study, Salani et al. tested 520 cytokinin derivatives and reported that 214 of these new compounds were more active than K. However, structures were shown only for two of the active compounds – 2-chloro- 5. Protective effect of other cytokinins 8-((3,3-difluorocyclobutyl)methoxy)N-(thiazol-2-ylmethyl)-9H-purin- 6-amine and 2-chloro-8-(2-methoxyethoxy)-N-(pyrimidin-4-ylmethyl)- Reports of kinetin having protective activity in animal systems 9H-purin-6-amine (Salani et al., 2019). prompted interest in the effects of other cytokinins and their synthetic Protective activity has also been reported for some cytokinin ribo- derivatives. Some of their activities have recently been reviewed else- sides. These compounds are typically cytotoxic at higher concentra- where (Voller et al., 2017). tions, which is why most research has focused on their potential ap- Natural compounds such as trans-zeatin (tZ, Fig. 2A), N6-iso- plication as cancer chemotherapeutic drugs. Nevertheless, N6- pentenyladenine (iP, Fig. 2B), N6-benzyladenine (BA, Fig. 2C) and para- isopentenyladenosine, N6-benzyladenosine (Dassano et al., 2014) and topolin (Fig. 2D) were reported to have antioxidant activity in vitro other cytokinin ribosides, with the notable exception of 6-(2-hydroxy-3- (Choi et al., 2009; Jabłonska-Trypuc et al., 2016; Brizzolari et al., methoxybenzyl)adenosine (unpublished), are activators of the Nrf2 2016). pathway. Ortho-topolin riboside (Huang et al., 2011), trans-zeatin ri- tZ is another cytokinin used as an active ingredient in cosmetics, boside, and possibly also kinetin riboside (Lee et al., 2012) have been although as far as we know, results of clinical tests of its safety and proposed to act as agonists of adenosine A2A receptors, which are a efficacy in humans have not yet been published. Nevertheless, tZ was potential target of multiple neurodegenerative and other diseases able to modulate the aging of fibroblasts in a manner similar to K (Pisanti et al., 2014). (Rattan and Sodagam, 2005), and it protected skin cells against UV irradiation, increased the expression of aquaporin 3 and improved the 6. Conclusions healing/migration capability of cells in a scratch assay (Ji et al., 2010; Yang et al., 2009). Some studies have also shown that tZ has neuro- N6-furfuryladenine, otherwise known as kinetin (K), is a well-known protective activity – it protected the rat pheochromocytoma cell line member of the cytokinins, a class of plant hormones. Although widely PC12 against toxicity induced by β-amyloid treatment and improved considered to be a natural molecule, present not only in plants but in cognitive functions in mice with scopolamine-induced amnesia (Choi animals as well, inconsistencies in studies focused on its detection in et al., 2009; Kim et al., 2008). various materials have made us wonder under which circumstances it Both ortho- and para-topolin prolonged the lifespan of C. elegans really occurs in nature. Apart from a variety of activities in plants, K has (Kadlecová et al., 2018). Moreover, a recent study showed that topical also been shown to have a health-promoting effect in animal systems. It application of the latter compound improved the appearance of pho- was able to modulate aging in vitro and in invertebrates and it pre- todamaged skin in humans (Garcia et al., 2018). vented deterioration in multiple tissues in rodent models of accelerated Cytokinins may also find applications in the treatment of skin hy- aging. What is lacking, however, are data describing its effect on nat- popigmentation disorders such as vitiligo. In B16 melanoma cells, BA ural aging in mammals, as well as studies focused on an increase in

8 A. Kadlecová, et al. Ageing Research Reviews xxx (xxxx) xxxx

Fig. 3. Protective effect of cytokinin derivatives. A) 3-day pre-treatment of C. elegans with cytokinin derivatives protects against/sensitizes to oxidative (500 μM juglone, 5 h) and heat (35 °C, 6 h) stress. The set tested is enriched for compounds with protective activity, the majority of which have a substitution in position 8 of the purine ring. B) A 6 -h pre-treatment with 2 screening hits partially protects fibroblasts derived from a patient with Friedreich ataxia against stress caused by 48 -h exposure to 15 μM of the inhibitor of glutathione synthesis L-buthionine sulfoximine in a resazurin assay. healthspan – a disease free period preceding an age-related physiolo- ingredient in cosmetics for years and it has generally been reported as gical decline – rather than lifespan itself. We would argue that inclusion being safe and well tolerated after topical application. It is also in phase of K in programs such as the Interventions Testing Program could be of 2 of clinical trials for treatment of familial dysautonomia. Results from benefit1 . K has also been reported to possess some protective activity in pre-clinical studies and from initial clinical evaluation seem to be en- neurodegenerative, cardiovascular and metabolic diseases in vitro and couraging so far. K has even been detected in mouse brain tissue, which in vivo and it corrects splicing of the ELP-1 gene, mutations in which shows it can cross the blood brain barrier. In patients, short-term oral cause familial dysautonomia. administration of K did not cause any serious adverse effects. It remains K has been reported to induce antioxidant defenses and directly to be seen whether this promising trend will be maintained in long-term quench ROS. It modulates alternative splicing of certain pre-mRNAs, studies, especially given the presence of the furan moiety, which has most likely by increasing recognition of the 5’ splice site. In cells, K can been associated with hepatotoxicity. However, taking into account the be transformed into KR, KRMP, KRDP and KRTP. The latter was re- protective effect of K shown in various assays, further stages of clinical ported to act as a neo-substrate of the kinases PINK1 and casein kinase trials and the subsequent use of K for the treatment of FD or other 2, boosting their activity. The base, the riboside, or other products of approved indications may also provide information about its health- K´s metabolism, may also induce adaptive hormetic responses. It is promoting "side effects" in humans. unclear at this point whether, and if so how, these various activities are It is clear that we still need more data to even begin to understand connected and whether the effects observed in various organisms and the precise mechanism of kinetin´s action. Detailed evaluation of its tissues, although sometimes similar in nature, result from a shared effects in various tissues and organisms could be beneficial. Still, mechanism of action. hopefully it will not prove out of place to allow ourselves a little op- These promising results have also led to an interest in creating more timism, and hope that in the near future, K could bring much-needed active derivatives of K. Examples include the development of ProTide relief to patients suffering from certain debilitating and currently in- prodrugs aimed at delivering KRMP to cells, the preparation of more curable disorders. active modulators of splicing or our own projects aimed at developing novel cytokinin-inspired molecules for treating skin and metabolic Acknowledgements diseases. K may be a multitarget molecule, and it is therefore possible that focusing on the improvement of a given activity would lead to the This study was supported by the Ministry of Education, Youth and loss of other beneficial effects. A particularly interesting question is Sports of the Czech Republic (INTER-COST LTC17072, INTER-COST whether the reactivity of the furan ring is important for K´s protective LTC18078) and by the European Regional Development Fund - Project activity, since finding a bioisosteric replacement would be challenging. ENOCH (No. CZ.02.1.01/0.0/0.0/16_019/0000868). The authors are Nevertheless, such attempts are logical, as the activity of K is rather grateful to Sees-editing for language correction. AK stay in M. A-S. lab low. The fact that typically concentrations of tens or even hundreds of was supported by a Short Term Scientific Mission financed by GENiE μM of K are required in in vitro assays makes it challenging to provide (COST Action BM1408) and by a European Research Council grant effective doses in vivo. (ERC-2011-StG-281691) to M. A.-S. The Friedreich ataxia related stu- On the other hand, the toxicity of K seems to be remarkably low as dies were supported by COST (European Cooperation in Science and well, as indicated by the limited cytotoxicity it displays even after long- Technology) Action CA15133 (FeSBioNet). term cultivation experiments. Moreover, K has been used as an active References

1 https://www.nia.nih.gov/research/dab/interventions-testing-program-itp Amasino, R., 2005. 1955: Kinetin arrives. The 50th anniversary of a new plant hormone.

9 A. Kadlecová, et al. Ageing Research Reviews xxx (xxxx) xxxx

Plant Physiol. 138 (3), 1177–1184. Grace, J., Hakala, M., RM H, J.B., 1967. N6-substituted adenine derivatives as growth An, S., Cha, H.J., Ko, J.M., Han, H., Kim, S.Y., Kim, K.S., Lee, S.J., An, I.S., Kim, S., Youn, inhibitors of human leukemic myeloblasts and S-180 cells. Proc. Am. Assoc. Cancer H.J., et al., 2017. Kinetin improves barrier function of the skin by modulating ker- Res. 8, 23–27. atinocyte differentiation markers. Ann. Dermatol. 29 (1), 6–12. Gu, X., Greiner, E.R., Mishra, R., Kodali, R., Osmand, A., Finkbeiner, S., Steffan, J.S., Atwal, R.S., Desmond, C.R., Caron, N., Maiuri, T., Xia, J., Sipione, S., Truant, R., 2011. Thompson, L.M., Wetzel, R., Yang, X.W., 2009. Serines 13 and 16 are critical de- Kinase inhibitors modulate huntingtin cell localization and toxicity. Nat. Chem. Biol. terminants of full-length human mutant huntingtin induced disease pathogenesis in 7 (7), 453. HD mice. Neuron 64 (6), 828–840. Axelrod, F.B., Liebes, L., Gold-von Simson, G., Mendoza, S., Mull, J., Leyne, M., Norcliffe- Hall, R.H., De Ropp, R., 1955. Formation of 6-furfurylaminopurine from DNA breakdown Kaufmann, L., Kaufmann, H., Slaugenhaupt, S.A., 2011. Kinetin improves IKBKAP products. J. Am. Chem. Soc. 77 (23), 6400. mRNA splicing in patients with familial dysautonomia. Pediatr. Res. 70 (5), 480. Hampton, A., Biesele, J.J., Moore, A.E., Brown, G.B., 1956. 6-furfurylamino-9-β-dribo- Bakhiya, N., Appel, K.E., 2010. Toxicity and carcinogenicity of furan in human diet. Arch. furanosylpurine: synthesis and differential toxicity to mammalian cells in vitro. J. Am. Toxicol. 84 (7), 563–578. Chem. Soc. 78 (21), 5695. Barciszewski, J., Siboska, G.E., Pedersen, B.O., Clark, B.F., Rattan, S.I., 1996. Evidence for Harman, D., 1955. Aging: a theory based on free radical and radiation chemistry. J. the presence of kinetin in DNA and cell extracts. FEBS Lett. 393 (2–3), 197–200. Gerontol. 11 (3), 298–300. Barciszewski, J., Siboska, G.E., Pedersen, B.O., Clark, B.F., Rattan, S.I., 1997a. Furfural, a Hertz, N.T., Berthet, A., Sos, M.L., Thorn, K.S., Burlingame, A.L., Nakamura, K., Shokat, precursor of the cytokinin hormone kinetin, and base propenals are formed by hy- K.M., 2013. A neo-substrate that amplifies catalytic activity of Parkinson’s-disease- droxyl radical damage of DNA. Biochem. Biophys. Res. Commun. 238 (2), 317–319. related kinase PINK1. Cell 154 (4), 737–747. Barciszewski, J., Siboska, G.E., Pedersen, B.O., Clark, B.F., Rattan, S.I., 1997b. A me- Hims, M.M., Leyne, M., Mull, J., Liu, L., Lazaro, C., Shetty, R.S., Gill, S., Gusella, J.F., chanism for the in vivo formation of N6-furfuryladenine, kinetin, as a secondary Reed, R., Slaugenhaupt, S.A., et al., 2007. Therapeutic potential and mechanism of oxidative damage product of DNA. FEBS Lett. 414 (2), 457–460. kinetin as a treatment for the human splicing disease familial dysautonomia. J. Mol. Barciszewski, J., Rattan, S.I., Siboska, G., Clark, B.F., 1999. Kinetin—45 years on. Plant Med. 85 (2), 149–161. Sci. 148 (1), 37–45. Hönig, M., Plíhalová, L., Spíchal, L., Grúz, J., Kadlecová, A., Voller, J., Svobodová, A.R., Barciszewski, J., Mielcarek, M., Stobiecki, M., Siboska, G., Clark, B.F., 2000. Vostálová, J., Ulrichová, J., Doležal, K., et al., 2018. New cytokinin derivatives Identification of 6-furfuryladenine (kinetin) in human urine. Biochem. Biophys. Res. possess UVA and UVB photoprotective effect on human skin cells and prevent oxi- Commun. 279 (1), 69–73. dative stress. Eur. J. Med. Chem. 150, 946–957. Berge, U., Kristensen, P., Rattan, S.I., 2006. Kinetin-induced differentiation of normal Honma, Y., Ishii, Y., 2002. Differentiation of human myeloid leukemia cells by plant human keratinocytes undergoing aging in vitro. Ann. N. Y. Acad. Sci. 1067 (1), redifferentiation-inducing hormones. Leuk. Lymphoma 43 (9), 1729–1735. 332–336. Hoydonckx, H., Van Rhijn, W., Van Rhijn, W., De Vos, D., Jacobs, P., 2000. Furfural and Berge, U., Kristensen, P., Rattan, S.I., 2008. Hormetic modulation of differentiation of Derivatives. Ullmann’s Encyclopedia of Industrial Chemistry. normal human epidermal keratinocytes undergoing replicative senescence in vitro. Hsiao, G., Shen, M.Y., Lin, K.H., Chou, C.Y., Tzu, N.H., Lin, C.H., Chou, D.S., Chen, T.F., Exp. Gerontol. 43 (7), 658–662. Sheu, J.R., 2003. Inhibitory activity of kinetin on free radical formation of activated Bolund L, Jensen PK, Bjerring P (1991) Method and composition for treating hyperpro- platelets in vitro and on thrombus formation in vivo. Eur. J. Pharmacol. 465 (3), liferative skin diseases using 6-aminopurine cytokinins. US Patent 5,021,422. 281–287. Boone, N., Loriod, B., Bergon, A., Sbai, O., Formisano-Tréziny, C., Gabert, J., Huang, N.K., Lin, J.H., Lin, J.T., Lin, C.I., Liu, E.M., Lin, C.J., Chen, W.P., Shen, Y.C., Khrestchatisky, M., Nguyen, C., Féron, F., Axelrod, F.B., et al., 2010. Olfactory stem Chen, H.M., Chen, J.B., et al., 2011. A new drug design targeting the adenosinergic cells, a new cellular model for studying molecular mechanisms underlying familial system for Huntington’s disease. PLoS One 6 (6), e20934. dysautonomia. PLoS One 5 (12), e15590. Ishii, Y., Hori, Y., Sakai, S., Honma, Y., 2002. Control of differentiation and apoptosis of Boone, N., Bergon, A., Loriod, B., Devèze, A., Nguyen, C., Axelrod, F.B., Ibrahim, E.C., human myeloid leukemia cells by cytokinins and cytokinin nucleosides, plant re- 2012. Genome-wide analysis of familial dysautonomia and kinetin target genes with differentiation inducing hormones. Cell Growth Differ. Publ. Am. Assoc. Cancer Res. patient olfactory ectomesenchymal stem cells. Hum. Mutat. 33 (3), 530–540. 13 (1), 19–26. Bowie, L.E., Maiuri, T., Alpaugh, M., Gabriel, M., Arbez, N., Galleguillos, D., Hung, C.L., Ishii, Y., Sakai, S., Honma, Y., 2003. Cytokinin-induced differentiation of human myeloid Patel, S., Xia, J., Hertz, N.T., et al., 2018. N6-furfuryladenine is protective in leukemia hl-60 cells is associated with the formation of nucleotides, but not with Huntington’s disease models by signaling huntingtin phosphorylation. Proc. Natl. incorporation into dna or rna. Biochimica et Biophysica Acta (BBA)-Molecular Cell Acad. Sci. 115 (30), E7081–E7090. Research 1643 (1–3), 11–24. Brizzolari, A., Marinello, C., Carini, M., Santaniello, E., Biondi, P.A., 2016. Evaluation of Jabłonska-Trypuc, A., Matejczyk, M., Czerpak, R., 2016. N6-benzyladenine and kinetin the antioxidant activity and capacity of some natural N6-substituted adenine deri- influence antioxidative stress parameters in human skin fibroblastst. Mol. Cell. vatives (cytokinins) by fluorimetric and spectrophotometric assays. J. Chromatogr. B Biochem. 413 (1–2), 97–107. 1019, 164–168. Jeon, S.M., 2016. Regulation and function of AMPK in physiology and diseases. Exp. Mol. Cabello, C., Bair, W., Ley, S., Lamore, S., Azimian, S., Wondrak, G., 2009. The experi- Med. 48 (7), e245. mental chemotherapeutic N6-furfuryladenosine (kinetin-riboside) induces rapid ATP Ji, C., Yang, Y., Yang, B., Xia, J., Sun, W., Su, Z., Yu, L., Shan, S., He, S., Cheng, L., et al., depletion, genotoxic stress, and CDKN1A (p21) upregulation in human cancer cell 2010. Trans-zeatin attenuates ultraviolet induced down-regulation of aquaporin-3 in lines. Biochem. Pharmacol. 77 (7), 1125–1138. cultured human skin keratinocytes. Int. J. Mol. Med. 26 (2), 257–263. Yong, J., Cheong, J., Goh, D., Tan, S., Ong, E., 2009. Inhibitory effect of kinetin riboside Kadlecová, A., Jirsa, T., Novák, O., Kammenga, J., Strnad, M., Voller, J., 2018. Natural in human heptamoa, HepG2. Mol. Biosyst. 5 (1), 91–98. plant hormones cytokinins increase stress resistance and longevity of Caenorhabditis Chiu, P.C., Chan, C.C., Lin, H.M., Chiu, H.C., 2007. The clinical anti-aging effects of to- elegans. Biogerontology 19 (2), 109–120. pical kinetin and niacinamide in Asians: a randomized, double-blind, placebo-con- Kieber, J.J., Schaller, G.E., 2014. Cytokinins. The Arabidopsis Book/American Society of trolled, split-face comparative trial. J. Cosmet. Dermatol. 6 (4), 243–249. Plant Biologists 12. Choi, S.J., Jeong, C.H., Choi, S.G., Chun, J.Y., Kim, Y.J., Lee, J., Shin, D.H., Heo, H.J., Kim, M.J., Choi, S.J., Lim, S.T., Kim, H.K., Kim, Y.J., Yoon, H.G., Shin, D.H., 2008. Zeatin 2009. Zeatin prevents amyloid β-induced neurotoxicity and scopolamine-induced supplement improves scopolamine-induced memory impairment in mice. Biosci. cognitive deficits. J. Med. Food 12 (2), 271–277. Biotechnol. Biochem. 72 (2), 577–581. Dassano, A., Mancuso, M., Giardullo, P., De Cecco, L., Ciuffreda, P., Santaniello, E., Saran, Kim, S., Lee, J., Jung, E., Lee, J., Huh, S., Hwang, H., Kim, Y., Park, D., 2009. 6-benzy- A., Dragani, T.A., Colombo, F., 2014. N6-isopentenyladenosine and analogs activate laminopurine stimulates melanogenesis via cAMP-independent activation of protein the Nrf2-mediated antioxidant response. Redox Biol. 2, 580–589. kinase A. Arch. Dermatol. Res. 301 (3), 253–258. Davalli, P., Mitic, T., Caporali, A., Lauriola, A., D’Arca, D., 2016. ROS, cell senescence, Kimura, T., Doi, K., 2004. Depigmentation and rejuvenation effects of kinetin on the aged and novel molecular mechanisms in aging and age-related diseases. Oxid. Med. Cell. skin of hairless descendants of Mexican hairless dogs. Rejuvenation Res. 7 (1), 32–39. Longev. 2016. Kleiner, R.E., Kapoor, T.M., 2013. Bumping up kinase activity with an ATP-derived neo- Doležal, K., Popa, I., Hauserová, E., Spíchal, L., Chakrabarty, K., Novák, O., Kryštof, V., substrate. Cell 154 (4), 716–718. Voller, J., Holub, J., Strnad, M., 2007. Preparation, biological activity and en- Klinkenberg, M., Thurow, N., Gispert, S., Ricciardi, F., Eich, F., Prehn, J., Auburger, G., dogenous occurrence of N6-benzyladenosines. Bioorg. Med. Chem. 15 (11), Kögel, D., 2010. Enhanced vulnerability of PARK6 patient skin fibroblasts to apop- 3737–3747. tosis induced by proteasomal stress. Neuroscience 166 (2), 422–434. Donato, A.J., Morgan, R.G., Walker, A.E., Lesniewski, L.A., 2015. Cellular and molecular Kowalska, E., 1992. Influence of kinetin (6-furfurylo-amino-purine) on human fibroblasts biology of aging endothelial cells. J. Mol. Cell. Cardiol. 89, 122–135. in the cell culture. Folia Morphol. 51 (2), 109–118. Fleysher, M.H., Bloch, A., Hakala, M., Nichol, C., 1969. Synthesis and biological activity Lee, G., Papapetrou, E.P., Kim, H., Chambers, S.M., Tomishima, M.J., Fasano, C.A., Ganat, of some new N6-substituted purine nucleosides. J. Med. Chem. 12 (6), 1056–1061. Y.M., Menon, J., Shimizu, F., Viale, A., et al., 2009. Modelling pathogenesis and Friedman, J., 2018. Neurofibromatosis 1. GeneReviews. PMID: 20301288. . treatment of familial dysautonomia using patient-specific iPSCs. Nature 461 (7262), Garcia, R., McCullough, J., Reece, B., Massino, F., 2018. Novel topical cytokinin improves 402. coarse wrinkles and skin roughness without skin irritation. J. Drugs Ermatol. 17 (7), Lee, J.H., Chung, K.Y., Bang, D., Lee, K.H., 2006. Searching for aging-related proteins in 772–779. human dermal microvascular endothelial cells treated with anti-aging agents. Ge, L., Yong, J.W.H., Goh, N.K., Chia, L.S., Tan, S.N., Ong, E.S., 2005. Identification of Proteomics 6 (4), 1351–1361. kinetin and kinetin riboside in coconut (Cocos nucifera l.) water using a combined Lee, Y.C., Yang, Y.C., Huang, C.L., Kuo, T.Y., Lin, J.H., Yang, D.M., Huang, N.K., 2012. approach of liquid chromatography–tandem mass spectrometry, high performance When cytokinin, a plant hormone, meets the adenosine A2A receptor: a novel neu- liquid chromatography and capillary electrophoresis. J. Chromatogr. B 829 (1–2), roprotectant and lead for treating neurodegenerative disorders? PLoS One 7 (6), 26–34. e38865. Goindi, S., Guleria, A., Aggarwal, N., 2015. Development and evaluation of solid lipid Lemke, R.A., Peterson, A.C., Ziegelhoffer, E.C., Westphall, M.S., Tjellström, H., Coon, J.J., nanoparticles of N6-furfuryl adenine for prevention of photoaging. J. Biomed. Donohue, T.J., 2014. Synthesis and scavenging role of furan fatty acids. Proc. Natl. Nanotechnol. 11 (10), 1734–1746. Acad. Sci. 111 (33), E3450–E3457.

10 A. Kadlecová, et al. Ageing Research Reviews xxx (xxxx) xxxx

Li, M., Ouyang, W., Wu, X., Zheng, Y., Gao, R., Ouyang, X., An, L., et al., 2014a. Kinetin Pratviel, G., Pitié, M., Bernadou, J., Meunier, B., 1991. Furfural as a marker of DNA ’ protects D-galactose-induced aging rats from splenic oxidative damage. Pak. Vet. J. cleavage by hydroxylation at the 5 carbon of deoxyribose. Angew. Chemie Int. Ed. 34, 386–390. English 30 (6), 702–704. Li, M., Ouyang, W., Wu, X., Zheng, Y., Gao, R., Tang, J., 2014b. Effect of kinetin on Pros, E., Fernández-Rodríguez, J., Benito, L., Ravella, A., Capellá, G., Blanco, I., Serra, E., immunity and splenic lymphocyte proliferation in vitro in D-galactose-induced aging Lázaro, C., 2010. Modulation of aberrant NF1 pre-mRNA splicing by kinetin treat- rats. Sheng li xue bao:[Acta physiologica Sinica] 66 (5), 605–611. ment. Eur. J. Hum. Genet. 18 (5), 614. Li, M., Ouyang, W., Wu, X., Zheng, Y., Wei, Y., An, L., 2014c. Kinetin inhibits apoptosis of Radhakrishna, V., Nanilu, S.K., Sanjeev, G., Shetty, J., Somyaji, Y.T., Moodithaya, S.S., aging spleen cells induced by D-galactose in rats. J. Vet. Sci. 15 (3), 353–359. 2017. Evaluation of the potency of kinetin on radiation induced behavioural changes Lidsky, T.I., 2014. Is the aluminum hypothesis dead? J. Occup. Environ. Med. 56 (5 in swiss albino mice. J. Clin. Diagn. Res. 11 (7), TF01. Suppl), S73. Raman, N., Elumalai, S., 1996. Presence of cytokinin in the root nodules of Casuarina Liguori, I., Russo, G., Curcio, F., Bulli, G., Aran, L., Della-Morte, D., Gargiulo, G., Testa, G., equisetifolia. Indian J. Exp. Biol. 34 (6), 577–579. Cacciatore, F., Bonaduce, D., et al., 2018. Oxidative stress, aging, and diseases. Clin. Rattan, S.I., 2002. N6-furfuryladenine (kinetin) as a potential anti-aging molecule. J. Anti. Interv. Aging 13 (757). Med. 5 (1), 113–116. Liu, Y., Zhang, Z., Yang, X., 2011. Kinetin protects against lipid peroxidation and im- Rattan, S.I., 2008. Principles and practice of hormetic treatment of aging and age-related proves antioxidant status in cultured astrocytes and mouse brain exposed to D-ga- diseases. Hum. Exp. Toxicol. 27 (2), 151–154. lactose. Afr. J. Biotechnol. 10 (55), 11721–11727. Rattan, S.I., Clark, B.F., 1994. Kinetin delays the onset of aging characteristics in human Ma, Q., 2013. Role of Nrf2 in oxidative stress and toxicity. Annu. Rev. Pharmacol. fibroblasts. Biochem. Biophys. Res. Commun. 201 (2), 665–672. Toxicol. 53, 401–426. Rattan, S.I., Sodagam, L., 2005. Gerontomodulatory and youth-preserving effects of Maia Campos, P.M.B.G., de Camargo Júnior, F.B., de Andrade, J.P., Gaspar, L.R., 2012. zeatin on human skin fibroblasts undergoing aging in vitro. Rejuvenation Res. 8 (1), Efficacy of cosmetic formulations containing dispersion of liposome with magnesium 46–57. ascorbyl phosphate, alpha-lipoic acid and kinetin. Photochem. Photobiol. 88 (3), Sadigh-Eteghad, S., Majdi, A., McCann, S.K., Mahmoudi, J., Vafaee, M.S., Macleod, M.R., 748–752. 2017. Dgalactose-induced brain ageing model: a systematic review and meta-analysis Maiuri, T., Bowie, L.E., Truant, R., 2018. DNA repair signaling of huntingtin: the next link on cognitive outcomes and oxidative stress indices. PLoS One 12 (8), e0184122. between late-onset neurodegenerative disease and oxidative DNA damage. DNA Cell Salani, M., Urbina, F., Brenner, A., Morini, E., Shetty, R., Gallagher, C.S., Law, E.A., Biol. Sunshine, S., Finneran, D.J., Johnson, G., et al., 2019. Development of a screening Mattson, M., 2008. Hormesis defined. Ageing Res. Rev. 7 (1), 1–7. platform to identify small molecules that modify ELP1 pre-mRNA splicing in familial McCullough, J., Weinstein, G., 2002. Clinical study of safety and efficacy of using topical dysautonomia. SLAS DISCOVERY: Adv. Life Sci. R&D 24 (1), 57–67. kinetin 0.1% (kinerase®) to treat photodamaged skin. Cosmet. Dermatol. 15, 29–32. Sharma, S.P., Kaur, P., Rattan, S.I., 1995. ) Plant growth hormone kinetin delays aging, McCullough, J.L., Garcia, R.L., Reece, B., 2008. A clinical study of topical Pyratine 6 for prolongs the lifespan, and slows down development of the fruitfly Zaprionus improving the appearance of photodamaged skin. J. Drugs Dermatol. 7 (2), 131–135. Paravittiger. Biochem. Biophys. Res. Commun. 216 (3), 1067–1071. McDaniel, D., Neudecker, B., DiNardo, J., Lewis, J., Maibach, H., 2005. Idebenone: a new Sharma, S.P., Kaur, J., Rattan, S.I., 1997. Increased longevity of kinetin-fed Zaprionus antioxidant–part I. Relative assessment of oxidative stress protection capacity com- fruitflies is accompanied by their reduced fecundity and enhanced catalase activity. pared to commonly known antioxidants. J. Cosmet. Dermatol. 4 (1), 10–17. IUBMB Life 41 (5), 869–875. Mijikovic D, Hranisavljevic J, Pietrzkowski Z (2007) Pharmaceutical compositions and Shetty, R.S., Gallagher, C.S., Chen, Y.T., Hims, M.M., Mull, J., Leyne, M., Pickel, J., Kwok, methods for metabolic modulation. US Patent App. 10/567,875. D., Slaugenhaupt, S.A., 2011. Specific correction of a splice defect in brain by nu- Miller, C.O., Skoog, F., Okumura, F.S., Von Saltza, M.H., Strong, F., 1955a. Structure and tritional supplementation. Hum. Mol. Genet. 20 (21), 4093–4101. synthesis of kinetin. J. Am. Chem. Soc. 77 (9), 2662–2663. Sheu, J.R., Hsiao, G., Shen, M., Chou, C., Lin, C., Chen, T.F., Chou, D., 2003. Inhibitory Miller, C.O., Skoog, F., Von Saltza, M.H., Strong, F., 1955b. Kinetin, a cell division factor mechanisms of kinetin, a plant growth-promoting hormone, in platelet aggregation. from deoxyribonucleic acid. J. Am. Chem. Soc. 77 (5), 1392. Platelets 14 (3), 189–196. Miller, C.O., Skoog, F., Okumura, F., Von Saltza, M., Strong, F., 1956. Isolation, structure Shohat, M., Hubshman, M.W., 2014. Familial dysautonomia. GeneReviews. PMID: and synthesis of kinetin, a substance promoting cell division. J. Am. Chem. Soc. 78 20301359. . (7), 1375–1380. Gold-von Simson, G., Goldberg, J.D., Rolnitzky, L.M., Mull, J., Leyne, M., Voustianiouk, Mlejnek, P., Doležel, P., 2005. Apoptosis induced by N6-substituted derivatives of ade- A., Slaugenhaupt, S.A., Axelrod, F.B., 2009. Kinetin in familial dysautonomia carriers: nosine is related to intracellular accumulation of corresponding mononucleotides in implications for a new therapeutic strategy targeting mRNA splicing. Pediatr. Res. 65 HL-60 cells. Toxicology in vitro 19 (7), 985–990. (3), 341. Okada, Y., Okajima, H., 1998. Scavenging activity of furan derivatives against hydroxyl Sinha, R., Kim, Y.J., Nomakuchi, T., Sahashi, K., Hua, Y., Rigo, F., Bennett, C.F., Krainer, radical generated by fenton system. Yakugaku zasshi: Journal of the Pharmaceutical A.R., 2018. Antisense oligonucleotides correct the familial dysautonomia splicing Society of Japan 118 (6), 226–230. defect in IKBKAP transgenic mice. Nucleic Acids Res. 46 (10), 4833–4844. Okada, Y., Kaneko, M., Okajima, H., 1996. Hydroxyl radical scavenging activity of Skoog, F., 1994. A personal history of cytokinin and plant hormone research. In: Mok, naturally occurring furan fatty acids. Biol. Pharm. Bull. 19 (12), 1607–1610. D.W.S., Mokeds, M.C. (Eds.), Cytokinin: Chemistry, Activity and Function. CRC Press, Yamano, K., Okatsu, K., Sato, Y., Matsuda, N., Negishi, L., Takahashi, A., Yamagata, A., Boca Raton, pp. 1–14. Goto-Ito, S., Mishima, M., Ito, Y., et al., 2018. Structural insights into ubiquitin Strong, F., Skoog, F., Miller, C.O., 1965. Cytokinins. Science 148 (3669), 532–533. phosphorylation by PINK1. Sci. Rep. 8 (1), 10382. Slaugenhaupt, S.A., Mull, J., Leyne, M., Cuajungco, M.P., Gill, S.P., Hims, M.M., Quintero, Olsen, A., Siboska, G.E., Clark, B.F., Rattan, S.I., 1999. N6-furfuryladenine, kinetin, pro- F., Axelrod, F.B., Gusella, J.F., 2004. Rescue of a human mRNA splicing defect by the tects against fenton reaction-mediated oxidative damage to DNA. Biochem. Biophys. plant cytokinin kinetin. Hum. Mol. Genet. 13 (4), 429–436. Res. Commun. 265 (2), 499–502. Sun, J., Liu, Y., Cao, T., Ouyang, W., 2013. Eff ect of kinetin on ovary and uterus in D- Orr, A.L., Rutaganira, F.U., de Roulet, D., Huang, E.J., Hertz, N.T., Shokat, K.M., galactose induced female mouse model of aging. Sheng li xue bao:[Acta physiologica Nakamura, K., 2017. Long-term oral kinetin does not protect against α-synuclein- Sinica] 65 (4), 389–394. induced neurodegeneration in rodent models of Parkinson’s disease. Neurochem. Int. Thornfeldt, C.R., Rizer, R.L., 2016. Superior efficacy of an herbal-based cosmeceutical 109, 106–116. compared with common prescription and cosmetic antiaging therapies. J. Drugs Orr, M.F., McSwain, B., 1957. The effects of kinetin upon epithelium and fibroblasts in Dermatol. 15 (2), 218–223. tissue culture. Cancer 10 (3), 617–624. Tournas, J.A., Lin, F.H., Burch, J.A., Selim, M.A., Monteiro-Riviere, N.A., Zielinski, J.E., Orr, M.F., McSwain, B., 1960. The effect of kinetin, kinetin ribofuranoside and gibberellic Pinnell, S.R., 2006. Ubiquinone, idebenone, and kinetin provide ineffective photo- acid upon cultures of skin and mammary carcinoma and cystic disease. Cancer Res. protection to skin when compared to a topical antioxidant combination of vitamins C 20 (9), 1362–1364. and E with ferulic acid. J. Invest. Dermatol. 126 (5), 1185. Truitt, A., Ortiz, A., Elkeeb, L., Hindiyeh, R., Aquino, L., Tran, M., Weinstein, G., 2009. Tremaine, A.M., Ortiz, A., Elkeeb, L., Tran, M., Weinstein, G., 2010. Long-term efficacy Topical PRK 124 (0.125%) lotion for improving the signs and symptoms of rosacea. J. and safety of topical PRK 124 (0.125%) lotion (Pyratine-XR) in the treatment of mild- Drugs Dermatol. 8 (5), 459–462. to-moderate rosacea. J. Drugs Dermatol. 9 (6), 647–650. Osgerby, L., Lai, Y.C., Thornton, P.J., Amalfitano, J., Le Duff, C.S., Jabeen, I., Kadri, H., Verbeke, P., Siboska, G.E., Clark, B.F., Rattan, S.I., 2000. Kinetin inhibits protein oxida- Miccoli, A., Tucker, J.H., Muqit, M.M., et al., 2017. Kinetin riboside and its ProTides tion and glycoxidation in vitro. Biochem. Biophys. Res. Commun. 276 (3), activate the Parkinson’s disease associated PTEN-induced putative kinase 1 (PINK1) 1265–1270. independent of mitochondrial depolarization. J. Med. Chem. 60 (8), 3518–3524. Vicanova, J., Bouez, C., Lacroix, S., Lindmark, L., Damour, O., 2006. Epidermal and Othman, E.M., Naseem, M., Awad, E., Dandekar, T., Stopper, H., 2016. The plant hor- dermal characteristics in skin equivalent after systemic and topical application of skin mone cytokinin confers protection against oxidative stress in mammalian cells. PLoS care ingredients. Ann. N. Y. Acad. Sci. 1067 (1), 337–342. One 11 (12), e0168386. Voller, J., Zatloukal, M., Lenobel, R., Doležal, K., Béreš, T., Kryštof, V., Spíchal, L., Parvez, M., Birdsall, W., 1990. Structure of a copper (II) kinetin complex. Acta Niemann, P., Džubák, P., Hajdúch, M., et al., 2010. Anticancer activity of natural Crystallographica Section C 46 (10), 1775–1778. cytokinins: A structure–activity relationship study. Phytochemistry 71 (11–12), Perez, V.I., Bokov, A., Van Remmen, H., Mele, J., Ran, Q., Ikeno, Y., Richardson, A., 2009. 1350–1359. Is the oxidative stress theory of aging dead? Biochimica et Biophysica Acta (BBA)- Voller, J., Maková, B., Kadlecová, A., Gonzalez, G., Strnad, M., 2017. Plant hormone General Subjects 1790 (10), 1005–1014. cytokinins for modulating human aging and age-related diseases. Hormones in Peterson, L.A., 2012. Reactive metabolites in the biotransformation of molecules con- Ageing and Longevity. Springer, pp. 311–335. taining a furan ring. Chem. Res. Toxicol. 26 (1), 6–25. Voller, J., Béres, T., Zatloukal, M., Džubák, P., Hajdúch, M., Doležal, K., Schmulling, T., Pisanti, S., Picardi, P., Ciaglia, E., Margarucci, L., Ronca, R., Giacomini, A., Malfitano, Strnad, M., 2019. Anti-cancer activities of cytokinin ribosides. Phytochem. Rev. 1–13. A.M., Casapullo, A., Laezza, C., Gazzerro, P., et al., 2014. ) Antiangiogenic effects of Wanitphakdeedecha, R., Meeprathom, W., Manuskiatti, W., et al., 2015. Efficacy and N6-isopentenyladenosine, an endogenous isoprenoid end product, mediated by AMPK safety of 0.1% kinetin cream in the treatment of photoaging skin. Indian J. Dermatol. activation. Faseb J. 28 (3), 1132–1144. Venereol. Leprol. 81 (5), 547.

11 A. Kadlecová, et al. Ageing Research Reviews xxx (xxxx) xxxx

Wei, Y., Liu, D., Zheng, Y., Li, H., Hao, C., Ouyang, W., 2017. Protective effects of kinetin Biochimica et Biophysica Acta (BBA)-Gene Structure and Expression 1625 (3), against aluminum chloride and D-galactose induced cognitive impairment and oxi- 239–245. dative damage in mouse. Brain Res. Bull. 134, 262–272. Yang, B., Ji, C., Kang, J., Chen, W., Bi, Z., Wan, Y., 2009. Trans-zeatin inhibits UVB- Zheng, Y., Wei, Y., Liu, D., Hao, C., Li, H., Ouyang, W., 2018. Neuroprotective effects of induced matrix metalloproteinase-1 expression via MAP kinase signaling in human kinetin against glutamate-induced oxidative cytotoxicity in HT22 cells: involvement skin fibroblasts. Int. J. Mol. Med. 23 (4), 555–560. of Nrf2 and Heme Oxygenase-1. Neurotox. Res. 33 (4), 725–737. Yoshida, M., Kataoka, N., Miyauchi, K., Ohe, K., Iida, K., Yoshida, S., Nojima, T., Okuno, Wu, J., Weinstein, G., Kricorian, G., Kormeili, T., McCullough, J., 2007. Topical kinetin Y., Onogi, H., Usui, T., et al., 2015. Rectifier of aberrant mRNA splicing recovers 0.1% lotion for improving the signs and symptoms of rosacea. Clin. Exp. Dermatol. 32 tRNA modification in familial dysautonomia. Proc. Natl. Acad. Sci. 112 (9), (6), 693–695. 2764–2769. Wyszko, E., Barciszewska, M.Z., Markiewicz, M., Szymanski, M., Markiewicz, W.T., Clark, Zeltner, N., Fattahi, F., Dubois, N.C., Saurat, N., Lafaille, F., Shang, L., Zimmer, B., Tchieu, B.F., Barciszewski, J., 2003. “Action-at-a distance” of a new DNA oxidative damage J., Soliman, M.A., Lee, G., et al., 2016. Capturing the biology of disease severity in a product N6-furfuryladenine (kinetin) on template properties of modified DNA. PSC-based model of familial dysautonomia. Nat. Med. 22 (12), 1421.

12 7.3 Supplement 3

Hönig, M., Plíhalová, L., Spíchal, L., Grúz, J., Kadlecova, A., Voller, J., Svobodová, A.R., Vostálová, J., Ulrichová, J., Doležal, K., & Strnad, M. (2018). New cytokinin derivatives possess UVA and UVB photoprotective effect on human skin cells and prevent oxidative stress. European journal of medicinal chemistry, 150, 946-957.

69 European Journal of Medicinal Chemistry 150 (2018) 946e957

Contents lists available at ScienceDirect

European Journal of Medicinal Chemistry

journal homepage: http://www.elsevier.com/locate/ejmech

Research paper New cytokinin derivatives possess UVA and UVB photoprotective effect on human skin cells and prevent oxidative stress

* Martin Honig€ a, b, Lucie Plíhalova a, b, , Lukas Spíchal a,Jirí Grúz b, Alena Kadlecova b, Jirí Voller b, Alena Rajnochova Svobodova c, Jitka Vostalov a c, Jitka Ulrichova c, Karel Dolezal a, b, Miroslav Strnad b a Department of Chemical Biology and Genetics, Centre of the Region Hana for Biotechnological and Agricultural Research, Faculty of Science, Palacký University, Slechtitelu 27, CZ-783 71, Olomouc, Czech Republic b Laboratory of Growth Regulators, Centre of the Region Hana for Biotechnological and Agricultural Research, Palacký University & Institute of Experimental Botany ASCR, Slechtitelu 27, CZ-783 71, Olomouc, Czech Republic c Department of Medical Chemistry and Biochemistry, Faculty of Medicine and Dentistry, Palacký University Olomouc, Hnevotínska 3, 775 15, Olomouc, Czech Republic article info abstract

Article history: Eleven 6-furfurylaminopurine (kinetin, Kin) derivatives were synthesized to obtain biologically active Received 20 November 2017 compounds. The prepared compounds were characterized using 1H NMR, mass spectrometry combined Received in revised form with HPLC purity determination and elemental C, H, N analyses. The biological activity of new derivatives 22 February 2018 was tested on plant cells and tissues in cytokinin bioassays, such as tobacco callus, detached wheat leaf Accepted 15 March 2018 chlorophyll retention bioassay and Amaranthus bioassay. The selected compounds were subsequently Available online 21 March 2018 tested on normal human dermal ﬁbroblasts (NHDF) and keratinocyte cell lines (HaCaT) to exclude possible phototoxic effects and, on the other hand, to reveal possible UVA and UVB photoprotective Keywords: Kinetin derivatives activity. The protective antioxidant activity of the prepared cytokinin derivatives was further studied and Aromatic cytokinins compared to previously prepared antisenescent compound 6-furfurylamino-9-(tetrahydrofuran-2-yl) UVA/UVB photoprotectivity purine (Kin-THF) using induced oxidative stress (OS) on nematode Caenorhabditis elegans damaged by 5- Caenorhabditis elegans hydroxy-1,4-naphthoquinone (juglone), a generator of reactive oxygen species. The observed biological Oxidative stress activity was interpreted in relation to the structure of the prepared derivatives. The most potent oxidative stress protection of all the prepared compounds was shown by 6-(thiophen-2-ylmethylamino)- 9-(tetrahydrofuran-2-yl)purine (6) and 2-chloro-6-furfurylamino-9-(tetrahydrofuran-2-yl)purine (9) derivatives and the results were comparable to Kin-THF. Compounds 6 and 9 were able to signiﬁcantly protect human skin cells against UV radiation in vitro. Both the derivatives 6 and 9 showed higher protective activity in comparison to previously known structurally similar compounds Kin and Kin-THF. The obtained results are surprising due to the fact that the prepared compounds showed to be inactive in the ORAC assay which proved that the compounds did not act as direct antioxidants as they were unable to directly scavenge oxygen radicals. © 2018 Elsevier Masson SAS. All rights reserved.

Abbreviations: AAPH, 2,20-azobis(2-amidino-propane) dihydrochloride; ArCKs, aromatic cytokinins; BAP, 6-benzylamiopurine; BJ, human foreskin fibroblast cell line; 2- DRA, 2-deoxyribose degradation assay; GPX, glutathione peroxidase; GSR, glutathione reductase; CAT, catalase; GSH, glutathione; HaCaT, human immortalized keratinocyte cell line; HL-60, human promyelocytic cell line; HPLC, high performance liquid chromatography; Kin, kinetin, 6-furfurylaminopurine; LPO, lipid peroxidase; MDA, malondialdehyde; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; NGM, nematode growth medium; NR, neutral red; NRK, epithelial rat kidney cell line; NHDF, normal human dermal fibroblasts; NMR, nuclear magnetic resonance; ORAC, oxygen radical absorbance capacity assay; PCR, polymerase chain reaction; ROS, reactive oxygen species; SOD, superoxide dismutase; TFA, trifluoroacetic acid; TLC, thin layer chromatography. * Corresponding author. Department of Chemical Biology and Genetics, Centre of the Region Hana for Biotechnological and Agricultural Research, Faculty of Science, Palacký University, Slechtitelu 27, CZ-783 71, Olomouc, Czech Republic. E-mail address: [email protected] (L. Plíhalova). https://doi.org/10.1016/j.ejmech.2018.03.043 0223-5234/© 2018 Elsevier Masson SAS. All rights reserved. M. Honig€ et al. / European Journal of Medicinal Chemistry 150 (2018) 946e957 947

1. Introduction mixture showed free-radical scavenging properties but when vehiculated into a cosmetic formulation, it augmented the skin Cytokinins are phytohormones that are able to regulate cyto- barrier function against UV damage [26]. The mechanism of action kinesis, cell growth and differentiation, as well as leaf senescence of Kin in relation to photoprotection in human skin remains un- and many other aspects of plant life [1]. Aromatic cytokinins known and few statistically significant trials have been published to (ArCKs) such as kinetin (Kin) and 6-benzylaminopurine (BAP) date. possess an aromatic substituent at the C6 atom of the purine In this study, we prepared eleven Kin derivatives with modified moiety. Kin that was for the first time isolated from heated deox- or substituted furfuryl ring, in several cases combined with addi- yribonucleic acid preparations [2] and later on identified e.g. in the tional N9-substituent, to investigate the impact of structural endosperm liquid of fresh young coconut fruit [3], was described as changes on the protection of plant, animal and human cells and a multiactive molecule with various biological effects both in ani- tissues. The structure activity relationship of the prepared com- mal and plant cells [4]. In the presence of other plant hormone pounds was studied using several plant bioassays including tobacco auxin, Kin is able to induce cell division in plant tissue culture [2]. callus, chlorophyll retention (detached wheat leaves), and Amar- This aside, Kin N9-substituted ArCKs such as N9-ribosides [5], N9- anthus bioassays. The cytotoxicity of the prepared compounds tetrahydropyran-2-yls (THP), N9-tetrahydrofuran-2-yls (THF) [6] against two human skin lines (fibroblasts, BJ; keratinocytes, HaCaT) and N9-chlorobutyls [7], are reported to be exceptionally potent in was determined. The phototoxicity and UVA/UVB protective effect promoting chlorophyll retention in detached wheat leaf senescence of selected compounds was tested on normal human dermal fi- bioassay [3]. The ability of Kin as well as N9-halogenoethyl ArCK broblasts NHDF and HaCaT. derivatives to delay the senescence of excised wheat leaves in both Caenorhabditis elegans is an amenable model for biomedical dark and light conditions correlated with their ability to affect research, including screening of new molecules. It is easy and cheap membrane lipid peroxidation which is a typical symptom of plant to maintain in a laboratory while being compatible with high- senescence [8] and relates to the protection of plant cells against throughput approaches, which allows evaluating the effect of oxidative stress (OS) [8,9]. ArCKs are increasingly becoming a compounds on a whole-organism level [27,28]. The protection of subject of human interest because, in addition to the regulation of this nematode against OS was also studied as well as the oxygen plant functions, they can also influence human and mammalian radical absorbance capacity (ORAC) of selected active compounds. cells. ArCKs ribosides, such as Kin riboside or 6-(2- hydroxybenzylamino)purine riboside (ortho-topolin riboside, oTR) display high cytotoxic activity in various human cancer cell lines 2. Results and discussion [5,10,11]. Trisubstituted ArCKs analogs such as 2-(hydrox- yethylamino)-6-benzylamino-9-methylpurine (Olomoucine) or 6- 2.1. Organic syntheses (benzylamino)-2(R)-{[1-(hydroxymethyl)propyl]amino}-9- isopropylpurine (Roscovitine) were even described as cyclin- Eleven kinetin derivatives were prepared via nucleophilic sub- dependent kinase inhibitors and anti-tumor agents [12,13]. stitution of 6-chloropurine, 6-chloro-9-(tetrahydrofuran-2-yl)pu- Although ArCK N9-ribosides are cytotoxic, THP, THF and N9- rine- or 2,6-dichloro-9-(tetrahydrofuran-2-yl)purine with the chloroalkyl derivatives showed none or only marginal cytotoxicity appropriate amines (Scheme 1). The structures are shown in [21]. The antioxidant properties of Kin have been reported in both Table 1. The prepared compounds were characterized by C, H, and N in vitro and in vivo experiments [14e19] and the ability of Kin to elemental analysis, HPLC-PDA-MS, 1 H NMR and 13C NMR. stimulate antioxidant enzymes catalase (CAT), glutathione peroxi- Elemental analysis, melting point, ESI þ MS and HPLC purity data dase (GPX) and glutathione reductase (GSR) has been published are listed in Table 2, while 1H and 13C NMR spectral data are pro- [18]. Both Kin and BAP positively influenced glutathione (GSH) vided in the Experimental part and the spectral records are pro- content and decreased the level of lipid membrane peroxidation in vided in the Supplementary data. 6-Chloro-9-(tetrahydrofuran-2- human skin fibroblasts [18]. Low Kin doses (100 nM) reduced yl)purine was prepared using a slightly modified method apoptosis and protected human keratinocyte HaCaT and rat described in the literature [6]. 2,6-Dichloro-9-(tetrahydrofuran-2- epithelial kidney NRT cells as well as peripheral lymphocytes from yl)purine was prepared following the same procedure. We pre- OS mediated cell death [19]. Kin retarded aging and prolonged the pared C6-furfurylamino, C6-(thienylmethyl)amino, C6-[2-(3- lifespan of fed fruit flies and showed stimulatory effects on the thienyl)ethyl]amino, C6-(tetrahydrofuran-2-yl)methylamino- and activity of CAT during the development and adult life of fruit flies C6-(cyklopentylmethyl)aminopurine derivatives. In addition, we [15,16]. Kin elevates the activity of antioxidant enzymes superoxide modified seven compounds (5e11) with tetrahydrofuran-2-yl ring dismutase (SOD), CAT and GPX in rats [17]. at the N9 position of purine moiety. Three compounds (9e11)were Recently, a group of four CKs including Kin, BAP, 6-(4- also substituted by a chlorine atom in position C2. The preparation hydroxybenzylamino)purine (para-topolin, pT) and 6- of compound 1 has been described by De Roulet et al. (2015) [29] isopentenylaminopurine (iP) was tested for their antioxidant ac- using similar reaction conditions (R: Et3N, solvent: butanol, 12 h, tivity in oxygen radical absorbance capacity assay (ORAC). Among 100 C) but we used propanol instead of butanol and the reaction the tested CKs, Kin showed the highest activity in the concentration time was shortened to 4 h. Compound 2 was previously prepared range used (up to 1 mM) [20]. via reaction of 6-chloropurine with 2-thiophenemethylamine in 2- Kin and its derivative 6-furfurylamino-N9-(tetrahydropyran-2- methoxyethanol [30]. The approach used in this paper, allowed the yl)purine (Pyratine, Pyr) prepared in 2008 [21] were reported to use of a much lower amount of relatively expensive 2- delay age-related characteristic of human fibroblasts in vitro thiophenemethylamine but it was necessary to prolong the reac- ® [22e24]. Using reconstructed skin equivalent Mimeskin , a positive tion time. The preparation of compound 4 was described by De effect of topical Kin treatment on both epidermis and dermis for- Roulet et al. (2015) [29] as a multistep reaction of adenine with mation and development was found [25]. Pyr reduced symptoms cyclopentylmethanol. The utilization of cyclopentylmethylamine in and signs of photo damaged facial skin in vivo even more efficiently reaction with 6-chloropurine as described here allowed the prep- than Kin [21,24]. In 2012, Campos et al. [26] carried out a preclinical aration of compound 4 in a simple one step reaction. The prepa- study on the dispersion of liposome with magnesium ascorbyl ration of all derivatives is described in greater detail in the phosphate, alpha-lipoic acid and Kin and they found that the Experimental section. 948 M. Honig€ et al. / European Journal of Medicinal Chemistry 150 (2018) 946e957

Scheme 1. Reaction scheme for the synthesis of prepared 2,6- and 2,9-disubstituted and 2,6,9-trisubstituted purine derivatives. a) Et3N, propanol, 100 C, 4 h (1, 2, 5, 9, 10), 5 h (3, 4, 6, 8) and 6 h (7); b) EtOAc, CF3COOH, NH3, RT, 3.5 h.

2.2. Stability in acidic solution derivatives on cytokinin activity was observed in tobacco callus bioassay. All the tested compounds showed comparable or slightly The tetrahydrofuran-2-yl (THF) group has been commonly used higher activity than BAP that was used as a standard. The activity of in organic chemistry as a protective group, readily removable under new compounds in callus bioassay is also comparable to those acidic conditions [31]. As pH may vary in particular bioassays, we obtained for Kin and Kin-THP. The little effect of substituting an verified that THF substituted compounds are stable and do not oxygen atom in the furfuryl ring by sulphur (2) in the tobacco callus breakdown to a corresponding free base under chosen bioassay bioassay has been described [32]. However, we observed no conditions and we used a procedure described in the literature [6]. decrease in activity for C6-saturated ring derivatives (1, 4) in accord 6-(Tetrahydrofuran-2-ylmethylamino)-9-(tetrahydrofuran-2-yl) with other findings [33,34]. N9-substituted derivatives were purine (5) was chosen as the model compound. We performed generally less active in Amaranthus bioassay in comparison to BAP HPLC stability measurement in 10 4 M stock solution with pH (on average about 20% lower except for the virtually inactive decreasing from 7 to 3. The results showed pH and time-dependent compound 5) in agreement with other reports [6,8] and observed in release of free base, as determined by HPLC. These are given in the case of Kin-THF and 6-(tetrahydrofuran-2-ylmethylamino)-9- Table 3. The tested compound was stable at pH 7 and 6 even after (tetrahydrofuran-2-yl)purine (5). The biological activity of 5 in 24 h following sample preparation. It started to decompose after Amaranthus bioassay was 60% lower than the activity of its free base 1 h at pH 3 (5.7% of the free base release) and after 24 h at pH 5 (7.4% (1). On the other hand, 2-chloro-6-(thiophen-2-ylmethyl)-9- of the free base release). A significant breakdown of tested com- (tetrahydrofuran-2-yl)aminopurine (11) was active in this bioassay pound occurred after 24 h at pH 4 (55.7% of the free base) and and this was comparable to BAP. increased significantly at pH 3 (99.6% of the free base release). Since The presence of an oxygen atom in the furan ring showed this to the pH of the media in the performed bioassays varied between 6 be a critical structural motif for slowing the breakdown of chloro- and 7, we conclude that the prepared compounds did not disinte- phyll in detached wheat leaf senescence bioassay. The replacement grate in the used bioassays. We also tested the pH stability of two of an oxygen atom by sulphur or carbon resulted in a decrease of most biologically active compounds (6 and 9). Their stability in anti-senescent activity (2, 4 and 11). On the other hand, the satu- water solution in pH range 6e3 showed similar pattern to model ration of furan ring showed no such negative impact on the anti- compound 5. These compounds were therefore stable at pH higher senescence activity e in other words, compounds with a tetrahy- than 5 even after 24 h after sample preparation (Table 3). drofuran ring slowed the breakdown of chlorophyll (1, 5 and 10). High activity of N9-THF and N9-THP derivatives in promoting chlorophyll retention has been reported [7,21,32] and also observed 2.3. Cytokinin activity in bioassays in compounds 5 and 6. While the replacement of an oxygen atom in the Kin molecule by sulphur in compound 2 resulted in the The prepared derivatives were tested in three cytokinin bio- reduction of anti-senescent activity by 40%, following substitution assays (tobacco callus, detached wheat leaf senescence and Amar- by the THF group in the position N9 (6) increased the activity to the anthus bioassay) and the results are presented in Table 4. The level of BAP. The presence of a chlorine atom in purine molecule at bioassays described above were used to test cytokinin activity and the C2 atom did not affect the activity in leaf senescence bioassay to evaluate the structure and activity relationships of the prepared (10). On the other hand, C2-chlorine substitution lowered or compounds in comparison to published data. completely reduced the anti-senescent activity of compounds 9 The lowest effect of the structural changes in prepared M. Honig€ et al. / European Journal of Medicinal Chemistry 150 (2018) 946e957 949

Table 1 2.4. Protection of Caenorhabditis elegans against induced oxidative Structures of the prepared compounds and their abbreviations. stress

Protective activity of the compounds against chemically- induced oxidative stress was evaluated in Caenorhabditis elegans exposed to the lethal concentration of 5-hydroxy-1,4- naphthalenedione (juglone; 500 mM) [35]. Worms were pre- treated with test compounds (100 mM) 3 days prior to the juglone exposure. In the initial screening, the effect of the compounds on the survival of wild-type N2 worms was evaluated after 4 h. Promising results were obtained in the case of the compounds 6 and 9 (data not shown). In the follow-up experiments, those Compound R1 R2 R3 compounds were tested on larger populations and the viability was 1 HH evaluated hourly for 12e14 h. This was also carried out with the BA17 (fem-1) strain with temperature inducible sterility, in order to ensure that FUDR which was used to prevent the reproduction of wild-type worms, did not interfere with the activity of compounds. 2 HH The results were similar in both experimental settings - the compounds 6 and 9 protected the worms against the oxidative stress (Fig. 1, Supplementary data). In the experiment with BA17, the effect of the compounds was compared with earlier prepared com- 3 HH pound Kin-THF as well due to certain structural similarities, e.g. structural fragment in N9 atom of purine moiety with protecting furanyl group. Kin-THF was prepared and tested as the compound that is able to delay senescence and ﬁbroblast aging, as well as 4 HH another compound similar to Kin, 6-furfurylamino-9-(tetrahydropyran-2-yl)purine (Kin-THP) but the mode of action of the compound is still unknown even when clinically tested on human skin and is currently used in cosmetic [21,24]. Kin-THF is also able to 5 H delay the senescence inﬂuence membrane lipid peroxidation in plants as we published previously [8]. Kin-THF protected the worms against the oxidative stress comparably to both active compounds 6 and 9. 6 H

2.5. Evaluation of toxicity of prepared derivatives on human skin cells 7 H The cytotoxicity of the prepared compounds was evaluated in human diploid fibroblasts (BJ, ATTC) and keratinocyte cell lines (HaCaT) by resazurin reduction assay after 24 h. The assay is based on reduction resofurin into fluorescent resazurin by metabolically 8 H active cells. The test compounds were only marginally toxic or non- toxic (decrease in resazurin fluorescence < 10%), even at the highest concentration tested (50 mM, data not shown).

9 Cl 2.6. Phototoxicity of compounds 1, 6, 8, 9 and 10 on human skin cells

10 Cl Compounds 1, 6, 8, 9 and 10 were chosen out of all the prepared compounds for phototoxicity testing with the aim of covering all the modiﬁcations of Kin structure. The pre-treatment of NHDF and HaCaT with the tested compounds and the follow-up exposure of 11 Cl the samples to a non-toxic UVA dose did not result in any decrease in cell viability. Neutral red (NR) incorporation into both cell types and thus selected test compounds can be considered as non- phototoxic in the used concentration range (3.9e125 mM). The results using NHDF/HaCaT are shown in Supplementary data (Tables 1s and 2s). A well-known phototoxic compound, chlor- and 11, respectively. As described [32e34] the prolongation of the promazine (CPZ), used as a standard in the validated NRU photo- bridge between N6-substituent and N6-amino group decreased toxicity test, was also used here as a positive control. CPZ treatment cytokinin activity (tobacco callus and Amaranthus bioassay) (3). The and following exposure to UVA radiation clearly decreased the structural change also induced a complete loss of cytokinin activity viability of NHDF as well as HaCaT with IC50 of 25.5 ± 3.5 mM and in detached wheat leaf senescence bioassay. 36.1 ± 4.7 mM, respectively. 950 M. Honig€ et al. / European Journal of Medicinal Chemistry 150 (2018) 946e957

Table 2 Elemental analyses, melting points (mp) and ESI þ MS of the prepared compounds.

þ Compound Elemental analysis calc./measured mp [C] Yield [%] HPLC [%] ESI þ MS [M þ H ]

%C %H %N

1 54.8/54.7 6.0/6.1 31.9/31.9 189e191 66 >99 220 2 51.9/51.9 3.9/4.0 30.3/30.4 251e253 88 >99 232 3 53.8/53.6 4.5/4.5 28.6/28.5 240e242 72 >97 246 4 52.2/53.3 7.7/7.9 27.7/28.2 215e219 81 >98 218 5 aaa<22 76 >99 290 6 55.8/55.9 4.9/4.9 23.2/22.6 119e121 61 >98 302 7 57.1/57.0 5.4/5.4 22.2/22.0 99e104 46 >98 316 8 57.1/56.9 5.4/5.3 22.2/22.0 134e139 65 >98 316 9 52.6/51.9 4.4/4.4 21.9/20.8 144e146 57 >99 320 10 51.9/51.9 5.6/5.9 21.6/20.7 97e101 63 >97 324 11 50/50.8 4.2/4.3 20.9/20.6 158e161 56 >97 336 a: amorphous, not possible to determine.

Table 3 The pH stability of compounds 5, 6 and 9 in 10 4 M neutral and acidic water solutions. pH stability was measured 1 h (A) and 24 h (B) after sample preparation. The percentage of the released free base was determined by HPLC.

pH Compound peak area [%]

5 Free base 6 Free base 9 Free base

(A) 1 h after sample preparation 7 100 nd ee ee 6 100 nd 100 nd 100 nd 5 100 nd 100 nd 100 nd 4 100 nd 99.4 0.6 100 nd 3 94.3 5.7 94.6 5.4 98.8 1.2 (B) 24 h after sample preparation 7 100 nd ee ee 6 100 nd 100 nd 100 nd 5 92.6 7.4 94.2 5.8 92.9 7.1 4 44.3 55.7 52.1 47.9 51.7 48.3 Fig. 1. Effect of compounds 6, 9 and Kin-THF against oxidative stress in Caenorhabditis 3 0.4 99.6 0.8 99.2 0.9 99.1 elegans (BA17). P-values were adjusted for multiple testing using Bonferroni method. Comparisons with p-values < 0.05 were considered as statistically signiﬁcant. nd: not detected; -: not tested.

Table 4 Relative cytokinin bioassay activity of the prepared derivatives at the optimal concentration compared with the activity of 6-benzylaminopurine (BAP) (100% means 10 5 M BAP for the Amaranthus betacyanin bioassay, 10 4 M BAP in the case of the senescence bioassay and 10 5 M BAP for the tobacco callus bioassay).

Compound Tobacco callus bioassay Amaranthus caudatus betacyanin bioassay Wheat leaf senescence bioassay

Concentration [mol.l 1] Relative activity [%] Concentration [mol.l 1] Relative activity [%] Concentration [mol.l 1] Relative activity [%]

e e e Kin 10 5 101 ± 5105 68 ± 3104 98 ± 4 e e e Kin-THF 10 5 111 ± 7104 70 ± 8104 125 ± 11 e e e 1 10 5 110 ± 8104 109 ± 6104 114 ± 8 e e e 2 10 5 102 ± 2105 95 ± 3104 60 ± 12 e e 3 10 5 77 ± 12 10 4 80 ± 6 e na. e e e 4 10 5 100 ± 1104 97 ± 1104 71 ± 3 e e e 5 10 5 100 ± 7104 38 ± 5104 116 ± 9 e e e 6 10 5 105 ± 3104 81 ± 5104 102 ± 1 e e 7 10 5 99 ± 1 e nt. 10 4 110 ± 7 e e 8 10 5 103 ± 9 e nt. 10 4 75 ± 17 e e e 9 10 5 112 ± 9104 88 ± 5104 75 ± 12 e e e 10 10 5 104 ± 8104 83 ± 8104 111 ± 16 e e 11 10 5 108 ± 7104 101 ± 12 e na. nt: not tested, na:non-active, Kin-THF (6-furfurylamino-9-(tetrahydrofuran-2-yl)purin). Kin (6-furfurylaminopurine).

2.7. UVA and UVB photoprotection of human skin cells by photoprotective potential of our compounds with a well-known compounds 1, 6, 8, 9 and 10 standard in the field - rosmarinic acid (RA), a naturally occurring photoprotective substance [36,37]. Derivatives 1, 6, 8, 9 and 10 were chosen for photoprotectivity Firstly, NHDF and HaCaT skin cells were pre-incubated with testing on the basis of preliminary data and due to the fact that tested compounds and exposed to a cytotoxic dose of UVA radia- these derivatives illustrated an extent of performed structural de- tion. All selected compounds showed higher cell viability (amount rivatizations of Kin molecule. The published data on Kin photo- of incorporated NR) compared to DMSO (control). The results of protectivity is ambiguous and therefore we compared the NHDF photoprotection are presented in Fig. 2 (A) while the M. Honig€ et al. / European Journal of Medicinal Chemistry 150 (2018) 946e957 951 photoprotection of HaCaT results are given in the Supplementary 1 mM. The most active compound, RA, showed twelve times and data (Fig. 1sA, Fig. 1sB). thirteen times higher activity than both Kin and compound 8, Prepared tested derivatives showed higher or comparable respectively. The rest of the tested compounds (1, 6, 9 and 10) had photoprotective activity to RA. Compound 6 exhibited the highest no detectable activity as shown in Table 5. photoprotective effect in both cell models and in case of NHDF was fi signi cantly more active than RA in all tested concentrations. It was 3. Conclusion very potent even at the lowest tested concentration of 3.9 mM. At this concentration, compound 6 was more than six times and three We prepared and characterized eleven ArCK derivatives. While times more effective than RA on NHDF and HaCaT, respectively. The compounds 1, 2 and 4 were prepared according to methods second most effective was compound 9, especially in HaCaT. described in the literature, compounds 3, 5e11 were designed and fi Excluding compound 10, all compounds were signi cantly more prepared for the first time. We tested the cytokinin activity of the effective at the lowest tested concentration. prepared compounds in three cytokinin bioassays. The changes In the second set of experiments, cells were pre-incubated with made in Kin structure only slightly affected activity in tobacco tested compounds and exposed to a toxic dose of UVB radiation. callus bioassay in which the compounds were all comparably active Similarly, application of all compounds resulted in higher NR to the used standard BAP. In accordance with the literature, the N9 incorporation compared to DMSO (control) as shown in Fig. 2 (B).In purine atom substitution by the THF group reduced the biological the case of photoprotection against UVB, there was less difference activity of Kin derivatives (compounds 5, 6, 9 and 10)inAmaranthus between the activities of individual compounds than in the case of bioassay. The presence of an oxygen atom in the furan ring proves UVA photoprotection. Also only compounds 1 and 6 were found to to be a critical structural motif for maintaining of anti-senescence fi be signi cantly more active than RA at the highest tested concen- activity in detached wheat leaf senescence bioassay. The replace- tration of NHDF protection. Compound 10 that exhibited the lowest ment of an oxygen atom by a sulphur or carbon atom resulted in UVA photoprotective properties on HaCaT, had the most photo- reduced anti-senescent activity (compounds 2, 4, and 11). On the protective effect on HaCaT against UVB irradiation. This phenom- other hand, the saturation of the furan ring significantly increased enon may be linked to a different cytotoxic mechanism of UVA and the anti-senescence activity (compounds 1, 5, and 10). None of the UVB radiation, as reviewed by Svobodova and Vostalova[38]. selected compounds (1, 6, 8, 9 and 10) was phototoxic in the con- Compound 6 was the second most effective one. Both compounds 6 centration range of 3.9e125 mM used either on the NHDF or HaCaT. and 10 were more than twice as effective as RA at the lowest m Same derivatives were also used for testing UVA and UVB photo- concentration tested (3.9 M) on HaCaT. On NHDF, the protection of protective properties. Compound 6-(thiophen-2-ylmethylamino)- compounds 6 and 10 was comparable with RA. In addition to per- 9-(tetrahydrofuran-2-yl)purine (6) and 2-chloro-6-furfurylamino- formed experiments, we compared the photoprotective activity of 9-(tetrahydrofuran-2-yl)purine (9) showed the highest photo- the active compounds 6 and 9 with their structurally similar but protectivity against UVA radiation on both NHDF and HaCaT. In the already described compounds Kin and 6-furfurylamino-9-(tetra- case of UVB photoprotection, compounds 6 and 6-(2- hydrofuran-2-yl)purine (Kin-THF) on NHDF cells. We observed tetrahydrofuran-2-ylmethyl)aminopurine (1) were found to be higher UVA as well as UVB photoprotective activity of both the the most effective in NHDF, while 2-chloro-6- compounds 6 and 9 compared to Kin as well as Kin-THF on NHDF. tetrahydrofurfurylamino-9-(tetrahydrofuran-2-yl)purine (10) (Fig. 3). possessed high photoprotection of HaCaT. Compound 6 seems to be the most promising as it possesses both UVA and UVB photo- 2.8. Oxygen radical absorbance capacity (ORAC) protective ability, even at low concentration (3.9 mM). Besides, tested compounds 6 and 9 were both able to protect C. elegans The radical scavenging activity of compounds 1, 6, 8, 9 and 10 against OS in vivo. We compared the photoprotective activity of also tested for phototoxicity and photoprotection together with RA these two derivatives with Kin and Kin-THF and we found out that as a positive control, was determined by ORAC assay. The assay was the protective activity of newly prepared derivatives was compa- recently used for the evaluation of the antioxidant activity and rably higher. These results indicate antioxidant properties of com- capacity of some natural N6-substituted adenine derivatives [20] pounds 6 and 9. Although compounds 6 and 9 protected C. elegans within the frame of cytokinin group and it was stated that the and NHDF from oxidative and UV stress, they did not act as direct antioxidant activity of Kin is the highest up to concentrations of radical scavengers in ORAC assay. These data suggest that the

Fig. 2. Photoprotective effect of the prepared compounds 1, 6, 8, 9, 10 and positive control rosmarinic acid (RA) against UVA- (A) UVBe (B) induced damage to NHDF. The results shown in the figure represent the medians of four replicates; the error bars represent the boundaries of the first and third quartiles. To prove statistical significance the Mann- Whitney test was performed. Asterisks denote values that differ significantly from values of rosmarinic acid (Mann Whitney test; p < 0.05, n ¼ 4). 952 M. Honig€ et al. / European Journal of Medicinal Chemistry 150 (2018) 946e957

Fig. 3. Photoprotective effect of the newly prepared compounds 6, 9, Kin and Kin-THF against UVA- (A) UVBe (B) induced damage to NHDF. The results shown in the ﬁgure represent the average of 3 replicates; the error bars represent the standard deviation.

Table 5 (ESIþ) therefore molecular ions were recorded in their protonated þ Radical scavenging activity of selected prepared forms [M þ H] . 1H NMR spectra were measured on a Jeol 500 SS compounds. spectrometer operating at a temperature of 300 K and a frequency Compound ORAC (TE) of 500.13 MHz The samples were prepared by dissolving the com- RA 9.14 ± (0.32) pounds in DMSO-d6. Tetramethylsilane (TMS) was used as an in- Kin 0.71 ± (0.11) ternal standard. Thin-layer chromatography (TLC) was carried out 1 <0.05 using silica gel 60 WF254 plates (Merck). CHCl3/MeOH (9:1, v/v) or 6 <0.05 EtOAc/MeOH/NH3 (34:4:2, v/v) were used as mobile phase. Puriﬁ- 8 0.77 ± (0.11) 9 <0.05 cation via column chromatography was carried out using silica gel 10 <0.05 Davisil R LC60A 40e63 mm.

Data are expressed as Trolox equivalents (TE) mean ± SD (n ¼ 6). 4.2. Synthesis

4.2.1. 6-Chloro-9-(tetrahydrofuran-2-yl)purine (A) mechanism of photo- and nematode protection against OS is in- The compound was prepared according to a modified procedure direct and triggers other mechanisms than direct interaction with described in the literature [6]. 6-Chloropurine (10 g; 64.7 mmol) ROS and therefore there must exist an alternative mode of action and 2,3-dihydrofuran (7.3 mL; 97 mmol) were dissolved in EtOAc worth study and explaining the mode of protection caused by these (200 mL) and TFA (10 mL; 130 mmol) was added dropwise. The compounds. reaction mixture was stirred on ice at room temperature for 3.5 h and then neutralized by the appropriate amount of ammonia and 4. Experimental water (1:2). The ethyl acetate phase was washed with water, dried over Na2SO4 and concentrated in vacuo. A yellow oily compound 4.1. General procedure was obtained. The pure product was obtained after crystallization from ethanol at 20 C overnight. 1H NMR (500 MHz, DMSO-d6) All reagents were purchased from commercial suppliers and d ppm 1.95e2.05 (m, 1 H); 2.12e2.23 (m, 1 H); 2.37e2.44 (m, 1 H); used as received. Elemental analyses (C, H, N) were determined on 2.46e2.52 (m, 1 H); 3.90 (q, J ¼ 7.54 Hz, 1 H); 4.15 (td, J ¼ 7.72, an EA1112 Flash analyser (Thermo-Finnigan). The melting points 6.27 Hz, 1 H); 6.35 (dd, J ¼ 6.72, 3.67 Hz, 1 H); 8.75 (s, 1 H); 8.76 (s, ® (mp) were determined on SMP 30 (Stuart ) apparatus. The chro- 1 H). matographic purity and mass spectra of the prepared compounds were analysed using HPLC-PDA-MS method. Compounds (10 mlof 4.2.2. 2,6-Dichloro-9-(tetrahydrofuran-2-yl)purine (B) 3.10 5 M in 1% methanol) were injected onto a reverse-phased 2,6-Dichloropurine (10 g; 52.9 mmol) and 2,3-dihydrofuran column (Symmetry C18, 5 mm, 150 mm 2.1 mm; Waters, Mil- (6 mL; 79.4 mmol) were dissolved in EtOAc (200 mL) and TFA ford, MA, USA) incubated at 25 C. Solvent (A) consisted of 15 mM (8.1 mL; 105.8 mmol) was added dropwise on ice. The mixture was ammonium formate adjusted to pH 4.0. Solvent (B) consisted of refluxed for 3.5 h and then neutralized by the appropriate amount methanol. At flow-rate of 200 ml/min, following binary gradient was of ammonia and water (1:2). The ethyl acetate phase was washed used: 0 min, 10% B; 0e24 min; linear gradient to 90% B; 25e34 min; with water, dried over Na2SO4 and concentrated in vacuo. A yellow isocratic elution of 90% B; 35e45 min; linear gradient to 10% B using oily compound was obtained. The pure product was obtained after the Waters Alliance 2695 Separations Module (Waters, Manchester, crystallization from ethanol at 20 C overnight. 1H NMR UK). The effluent was introduced then to Waters 2996 PDA detector (500 MHz, DMSO-d6) d ppm 1.95e2.05 (m, 1 H); 2.07e2.17 (m, 1 H); (Waters, Manchester, UK) (scanning range 210e700 nm with 2.38e2.43 (m, 2 H); 3.90 (q, J ¼ 7.54 Hz, 1 H); 4.14 (td, J ¼ 7.87, 1.2 nm resolution) and a tandem mass analyser Q-Tof micro Mass 5.65 Hz, 1 H); 6.28e6.31 (m, 1 H); 8.78 (s, 1 H). Spectrometer (Waters, Manchester, UK) with an electrospray source (source temperature 120 C, desolvation temperature 4.2.3. 6-(2-Tetrahydrofuran-2-ylmethyl)aminopurine (1) 300 C, capillary voltage 3 kV). Nitrogen was used as well as cone 6-Chloropurine (0.5 g; 3.2 mmol), tetrahydrofurfurylamine gas (50 l/h) and desolvation gas (500 l/h). Data acquisition was (403 ml; 3.9 mmol) and triethylamine (Et3N) (2.5 mL; 16 mmol) performed in the full scan mode (50e1000 Da), scan time of 0.5 s were dissolved in propanol (30 mL). The reaction was refluxed for and cone voltage 20 V. Analyses were performed in positive mode 4 h. The mixture was concentrated in vacuo. The residue was M. Honig€ et al. / European Journal of Medicinal Chemistry 150 (2018) 946e957 953 dissolved in water and extracted into EtOAc, dried over Na2SO4 and next 2-thiophenemethylamine (23 ml; 0.23 mmol) was added and evaporated in vacuo to form the white crystalline product. 1H NMR the reaction mixture was further refluxed for 2 h. Reaction mixture (500 MHz, DMSO-d6) d ppm 1.57 (br. s., 1 H); 1.69e1.92 (m, 3 H); was concentrated in vacuo, the residue was dissolved in water and 3.35e3.53 (m, 2 H); 3.56 (d, J ¼ 6.42 Hz, 1 H); 3.73 (d, J ¼ 6.42 Hz, extracted into EtOAc, dried over Na2SO4, evaporated in vacuo and 1 H); 4.02 (br. s., 1 H); 7.46 (br. s., 1 H); 8.04 (br. s., 1 H); 8.13 (br. s., mixed with diethyl ether. The product precipitated from diethyl 1 H); 12.86 (br. s., 1 H). 13C NMR (126 MHz, DMSO) d 154.93, 152.83, ether. 1H NMR (500 MHz, DMSO-d6) d ppm 1.92e2.00 (m, 1 H); 149.97, 139.30, 119.26, 77.40, 67.60, 44.22, 29.08, 25.58. 2.12e2.22 (m, 1 H); 2.31e2.38 (m, 1 H); 2.38e2.43 (m, 1 H); 3.85 (td, J ¼ 7.68, 6.34 Hz, 1 H); 4.08 (td, J ¼ 7.68, 6.50 Hz, 1 H); 4.79 (br. s., 4.2.4. 6-(2-Thiophen-2-ylmethyl)aminopurine (2) 2 H); 6.21 (dd, J ¼ 6.88, 3.82 Hz, 1 H); 6.88 (dd, J ¼ 5.04,3.44 Hz, 1 H); 6-Chloropurine (1 g; 6.47 mmol), 2-thiophenemethylamine 6.97 (dd, J ¼ 3.40, 1.03 Hz, 1 H); 7.27 (dd, J ¼ 5.12, 1.22 Hz, 1 H); 8.23 13 (730 ml; 7.12 mmol) and Et3N (2.24 mL; 16.175 mmol) were dis- (s, 2 H); 8.36 (br. s., 1 H). C NMR (126 MHz, DMSO) d 154.45, solved in propanol (65 mL). The reaction mixture was refluxed for 152.86, 149.00, 143.58, 139.79, 127.06, 125.85, 125.24, 120.09, 84.83, 4 h. Solid yellowish product was obtained after crystallization from 69.16, 38.58, 31.62, 24.90. reaction mixture at 4 C overnight. 1H NMR (500 MHz, DMSO-d6) d ppm 4.81 (br. s., 2 H) 6.90 (br. s., 1 H) 6.98 (br. s., 1 H) 7.29 (br. s., 4.2.9. 6-(5-Methylthiophen-2-ylmethylamino)-9-(tetrahydrofuran- 1 H) 8.07 (br. s., 1 H) 8.19 (br. s., 2 H) 12.93 (br. s., 1 H). 13C NMR 2-yl)purine (7) (126 MHz, DMSO) d 154.32, 152.78, 150.15, 143.77, 139.55, 127.09, A (0.148 g; 0.66 mmol), 3-methylthiophene-2-methylamine 125.84, 125.23, 119.34, 38.60. (0.1 g; 0.79 mmol) and Et3N (0.46 mL; 3.3 mmol) were sequentially dissolved in propanol (10 mL). The mixture was refluxed for 4.2.5. 6-(2-Thiophen-2-ylethyl)aminopurine (3) 6 h under argon atmosphere. The mixture was concentrated in 6-Chloropurine (1 g; 6.47 mmol), 2-thiophenethylamine (910 ml; vacuo. The residue was dissolved in water and extracted into EtOAc. 7.76 mmol) and Et3N (4.5 mL; 32.35 mmol) were dissolved in The organic fraction was dried over Na2SO4 and evaporated in propanol (65 mL). The reaction mixture was refluxed for 5 h. vacuo. The product was obtained after precipitation from diethyl Yellowish solid product precipitated from water and filtrated off. 1H ether. 1H NMR (500 MHz, DMSO-d6) d ppm 1.93e2.00 (m, 1 H) NMR (500 MHz, DMSO-d6) d ppm 3.10 (t, J ¼ 7.34 Hz, 2 H) 3.69 (br. 2.13e2.19 (m, 1 H) 2.20 (s, 3 H) 2.31e2.43 (m, 2 H) 3.81e3.89 (m, s., 2 H) 6.89 (d, J ¼ 2.75 Hz, 1 H) 6.90e6.96 (m, 1 H) 7.29 (d, 1 H) 4.05e4.11 (m, 1 H) 4.71 (br. s., 2 H) 6.21 (dd, J ¼ 6.88, 3.82 Hz, J ¼ 5.20 Hz, 1 H) 7.74 (br. s., 1 H) 8.05 (s, 1 H) 8.17 (br. s., 1 H) 12.89 1 H) 6.75 (d, J ¼ 4.89 Hz, 1 H) 7.15 (d, J ¼ 5.20 Hz, 1 H) 8.21 (br. s., 1 H) (br. s., 1 H). 13C NMR (126 MHz, DMSO) d 154.39, 152.87, 150.43, 8.22 (s, 1 H) 8.31 (br. s., 1 H). 13C NMR (126 MHz, DMSO) d 154.39, 142.17, 139.62, 127.44, 125.62, 124.46, 118.55, 41.93, 29.81. 152.85, 148.94, 139.71, 136.79, 133.96, 130.22, 123.49, 120.07, 84.83, 69.17, 36.76, 31.64, 24.91, 13.98. 4.2.6. 6-Cyclopentylmethylaminopurine (4) 6-Chloropurine (0.325 g; 2.1 mmol), cyclopentanemethylamine 4.2.10. 6-(5-Hydroxymethylfuran-2-ylmethylamino)-9- (403 ml; 2.5 mmol) and Et3N (1.5 mL; 10.5 mmol) were dissolved in (tetrahydrofuran-2-yl)purine (8) propanol (20 mL). The reaction mixture was refluxed for 5 h and A (0.25 g; 1.1 mmol), 5-hydroxymethylfuran-2-ylmethylamine concentrated in vacuo. White solid was precipitated from water and (0.169 g; 1.3 mmol) and Et3N (0.464 mL; 3.3 mmol) were dissolved filtrated off. 1H NMR (500 MHz) d 12.85 (s, 1H), 8.13 (s, 1H), 8.05 (s, in propanol (10 mL). The reaction mixture was refluxed for 5 h. A 1H), 7.61 (s, 1H), 3.66 (partial overlap, 2H), 2.28e2.15 (m, 1H), crude, concentrated reaction mixture was mixed with water. 1.69e1.58 (m, 2H), 1.58e1.48 (m, 2H), 1.48e1.37 (m, 2H), 1.30e1.16 Precipitated solid white product was filtrated off. 1H NMR (m, 2H). 13C NMR (126 MHz, DMSO) d 154.56, 152.83, 150.66, 139.43, (500 MHz, CHLOROFORM-d) d ppm 2.08e2.15 (m, 2 H) 2.42e2.57 118.24, 45.12, 30.32, 25.30. (m, 2 H) 2.79 (br. s., 1 H) 4.05 (q, J ¼ 7.64 Hz, 1 H) 4.26 (dt, J ¼ 8.48, 6.46 Hz, 1 H) 4.55 (s, 2 H) 4.80 (br. s., 2 H) 6.18 (d, J ¼ 3.06 Hz, 1 H) 4.2.7. 6-(Tetrahydrofuran-2-ylmethylamino)-9-(tetrahydrofuran- 6.22 (d, J ¼ 3.06 Hz, 1 H) 6.27 (dd, J ¼ 6.27, 3.21 Hz, 1 H) 6.36e6.63 2-yl)purine (5) (m, 1 H) 7.86 (s, 1 H) 8.37 (br. s., 1 H). 13C NMR (126 MHz, DMSO) A (1 g; 4.46 mmol), tetrahydrofurfurylamine (554 ml; d 154.83, 154.60, 152.88, 152.67, 148.95, 139.77, 120.12, 108.20, 5.36 mmol) and Et3N (3.2 mL; 22.3 mmol) were sequentially dis- 107.72, 84.85, 69.18, 56.13, 37.10, 31.64, 24.90. solved in propanol (50 mL). The mixture was refluxed for 4 h and then concentrated in vacuo. The residue was dissolved in water and 4.2.11. 2-Chloro-6-furfurylamino-9-(tetrahydrofuran-2-yl)purine extracted into EtOAc using liquid-liquid continuous extractor (9) (24 h). EtOAc solution was dried over Na2SO4 and evaporated in B (0.5 g; 1.93 mmol), furfurylamine (204 ml; 2.31 mmol) and vacuo. The product was obtained after purification via column Et3N (1.35 mL; 9.65 mmol) were sequentially dissolved in propanol chromatography using EtOAc: MeOH: NH3 (34:1:1; v:v) mobile (25 mL). The mixture was refluxed for 4 h, concentrated in vacuo phase as eluent. 1H NMR (500 MHz, DMSO-d6), d ppm: 1.53e1.64 and mixed with water. A yellowish solid was filtrated off. The pure (m, 1 H); 1.69e1.89 (m, 3 H); 1.95e2.02 (m, 1 H); 2.11e2.23 (m, 1 H); product was acquired after recrystallization from CHCl3 and 2.30e2.44 (m, 2 H); 3.37e3.48 (m, 1 H); 3.48e3.53 (m, 1 H); ethanol. 1H NMR (500 MHz, DMSO-d6) d ppm 1.93e2.01 (m, 1 H); 3.53e3.60 (m, 1 H); 3.70e3.76 (m, 1 H); 3.81e3.90 (m, 1 H); 3.98 (q, 2.12 (dt, J ¼ 12.38, 7.41 Hz, 1 H); 2.31e2.38 (m, 2 H); 3.82e3.88 (m, J ¼ 7.03 Hz, 1 H); 4.09 (q, J ¼ 7.44 Hz, 1 H); 6.21 (dd, J ¼ 6.88, 3.82 Hz, 1 H); 4.07 (td, J ¼ 7.79, 6.11 Hz, 1 H); 4.57 (d, J ¼ 5.50 Hz, 2 H); 6.15 (t, 1 H); 7.63 (br. s., 1 H); 8.17 (br. s., 1 H); 8.21 (s, 1 H). 13C NMR J ¼ 5.20 Hz, 1 H); 6.21 (d, J ¼ 2.75 Hz, 1 H); 6.33 (br. s., 1 H); 7.51 (dd, (126 MHz, DMSO) d 155.02, 152.93, 148.73, 139.53, 120.01, 84.84, J ¼ 1.83, 0.92 Hz, 1 H); 8.25 (s, 1 H); 8.72 (br. s., 1 H). 13C NMR 77.34, 69.16, 67.59, 44.19, 31.64, 29.07, 25.55, 24.89. (126 MHz, DMSO) d 155.20, 153.49, 152.55, 149.87, 142.54, 140.21, 119.17, 111.04, 107.65, 84.98, 69.28, 37.16, 31.76, 24.66. 4.2.8. 6-(Thiophen-2-ylmethylamino)-9-(tetrahydrofuran-2-yl) purine (6) 4.2.12. 2-Chloro-6-tetrahydrofurfurylamino-9-(tetrahydrofuran-2- A (0.5 g; 2.23 mmol), 2-thiophenemethylamine (275 ml; yl)purine (10) 2.68 mmol) and Et3N (1.6 mL; 11.15 mmol) were sequentially dis- B (0.5 g; 1.93 mmol), tetrahydrofurfurylamine (240 ml; solved in propanol (25 mL). The mixture was refluxed for 3 h, then 2.31 mmol) and Et3N (1.35 mL; 9.65 mmol) were sequentially 954 M. Honig€ et al. / European Journal of Medicinal Chemistry 150 (2018) 946e957 dissolved in propanol (25 mL). The mixture was stirred under reflux betacyanin was determined [39]. for 4 h, concentrated in vacuo, and mixed with water. The pure In the wheat leaf senescence bioassay, the tip sections of the firs product was obtained from crude precipitate after recrystallization leaf of Triticum aestivum cv. Hereward seedlings were cultivated in from petroleum ether. 1H NMR (500 MHz, DMSO-d6) d ppm microtiter plates containing a cytokinin solution. After 4 days in 1.50e1.64 (m, 1 H); 1.68e1.91 (m, 4 H); 1.92e2.02 (m, 1 H); darkness leaves were removed and the chlorophyll retention was 2.07e2.19 (m, 1 H); 2.29e2.38 (m, 2 H); 3.44e3.52 (m, 1 H); determined [39]. 3.53e3.60 (m, 1 H); 3.65e3.74 (m, 1 H); 3.80e3.89 (m, 1 H); Tested cytokinin derivatives were dissolved in DMSO and 3.93e4.03 (m, 1 H); 4.03e4.10 (m, 1 H); 6.06e6.19 (m, 1 H); further diluted as required in the media used for each bioassay to 8.17e8.22 (m, 1 H); 8.22 (s, 1 H). 13C NMR (126 MHz, DMSO) the final concentration range 10 4 to 10 8 M. The final concentra- d 155.60, 153.60, 149.64, 139.96, 119.02, 84.93, 77.02, 69.25, 67.58, tion of DMSO in the media did not exceed 0.5%. Five replicates were 44.43, 31.74, 29.13, 25.51, 24.67. prepared for each compound concentration and the entire tests were repeated at least three times. From the data acquired in these 4.2.13. 2-Chloro-6-(thiophen-2-ylmethyl)-9-(tetrahydrofuran-2-yl) bioassays, the concentration inducing the strongest biological aminopurine (11) response was used to calculate the relative activity. The activity of B (0.5 g; 1.93 mmol), 2-thiophenmethylamine (238 ml; BAP (benzylaminopurine; active standard) at the optimal concen- 5 2.23 mmol) and Et3N (1.35 mL; 9.65 mmol) were sequentially dis- tration (10 M BAP for the Amaranthus betacyanin bioassay, solved in propanol (25 mL). The mixture was refluxed for 5 h. 10 4 M BAP in the case of the senescence bioassay and 10 5 M BAP Yellowish solid crystallized from the reaction mixture at 4 C for the tobacco callus bioassay) was set at 100 and the activities of overnight. Pure product was obtained after re-crystallization from the tested compounds were related to it at their optimal concen- hot ethanol. 1H NMR (500 MHz, DMSO-d6) d ppm 1.92e2.01 (m, trations. Optimal concentration for BAP in Amaranthus caudatus 1 H); 2.07e2.17 (m, 1 H); 2.29e2.39 (m, 2 H); 3.85 (q, J ¼ 7.34 Hz, bioassay is 10 5 M, while its activity falls down rapidly at the 1 H); 4.03e4.10 (m, 1 H); 4.72 (d, J ¼ 5.50 Hz, 2 H); 6.15 (t, concentration 10 4 M because it becomes toxic. However the ac- J ¼ 5.04 Hz,1 H); 6.90 (dd, J ¼ 4.74, 3.52 Hz,1 H); 6.98 (d, J ¼ 2.75 Hz, tivity of prepared compounds (excluding compound 2) grows 1 H); 7.31 (d, J ¼ 4.59 Hz, 1 H); 8.24 (s, 1 H); 8.87 (br. s., 1 H). 13C NMR steadily through the whole concentration range and compounds (126 MHz, DMSO) d 154.98, 153.48, 149.87, 142.42, 140.25, 127.13, are not toxic at the highest concentration, hence their highest ac- 126.41, 125.66, 119.13, 84.99, 69.28, 38.81, 31.75, 24.66. tivity is observed at 10 4 M concentration. That is the reason why the activity of different compounds was compared at different 4.3. pH stability of 6-(tetrahydrofuran-2-ylmethylamino)-9- (optimal) concentrations. (tetrahydrofuran-2-yl)purine (5) 4.5. Resazurin reduction assay on human skin cells pH stability evaluation was slightly modified and performed according to the literature [6]. The pH stability of compounds 5, 6 Resazurin reduction assay is a standard test of toxicity based on and 9 was analysed by HPLC-PDA (System Gold; Beckman In- the measurement of reduction of blue weakly fluorescent resazurin struments, Fullerton, CA, USA); analytes were detected at 270 nm into red highly fluorescent resofurin by metabolically active cells. using PDA detector (Beckman System Gold 168). Solution of tested Using the assay, the effects of *24 h treatments with several con- compound (10 2 M; DMSO) was prepared and diluted to 10 4 M centrations of the compounds (six fold dilution, maximal using McIlvaine buffer solution for the appropriate pH (2, 3, 4, 5, 6 concentration ¼ *50 mM) on viability of human skin fibroblasts BJ or 7). One hour after incubation at 25 C, 5 mL of the solution was and keratinocytes HaCaT were evaluated. The cell lines were ob- directly injected onto a reversed phase column (Symmetry C18; tained from the American Type Culture Collection, Manassas, VA, 5 mm, 150 2.1 mm; Waters, Milford, USA). At flow-rate of 0.3 mL/ USA the German Cancer Research Center (DKFZ), Heidelberg, Ger- min, the following binary gradient was used: 0 min, 90% B; many, respectively. About 5000 cells were seeded per well of a 96- 0e24 min; linear gradient to 10% A; 25e34 min; isocratic elution of well plate 24 h before the treatment. DMSO vehiculum was used as 10% A; 35e45 min; linear gradient to 90% A, where A was 15 mM a negative control. After 24 h, 10 concentrated solution of resa- formic acid adjusted to pH 4 with ammonium and B was 100% zurin in DMEM medium was added to the cells into the final con- methanol The HPLC measurement of the solutions was repeated centration of 100 mM. Fluorescence (ex ¼ 570 nm, em ¼ 610 nm) after a 24 h incubation at 25 C. was measured after 3 h incubation.

4.4. Evaluation of cytokinin activity 4.6. Phototoxicity and photoprotection assessment

Three standard cytokinin bioassays were performed as NHDF were isolated from skin tissue specimens obtained from described previously [39,40], based on the stimulation of tobacco healthy patients undergoing plastic surgery at the Department of callus growth, the retention of chlorophyll in excised wheat leaves Plastic and Aesthetic Surgery (University Hospital Olomouc). The and the dark induction of betacyanin synthesis in Amaranthus use of skin tissue was in accordance with the Ethics Committee of cotyledons and used to reveal cytokinin activity as described [39]. the University Hospital and Faculty of Medicine and Dentistry, Tobacco callus bioassay was slightly modified according to Palacký University, Olomouc and all patients signed written Nisler et al. (2016) [40]. It was performed in 6-well microplates informed consent. The tissue was cut into pieces of approximately (3 mL of MS medium containing tested cytokinin in each well, into 1 1 cm, placed in Petri dishes and cultured in the mixture of which 0.1 g of callus was placed). Biological activity was deter- DMEM and Ham's F12 Nutrient Mixture (1:3) supplemented with mined from the increase in fresh callus weight after four weeks of FCS (10%; v/v), penicillin (100 mg/mL), streptomycin (100 U/mL), cultivation [39,40]. amphotericin B (0.125 mg/mL), hydrocortisone (0.8 mg/mL), In the Amarathus bioassay the seeds of Amaranthus caudatus var. adenine (24 mg/mL), insulin (0.12 U/mL), epidermal growth factor atropurpurea were cultivated for 3 days in darkness before the roots (1 ng/mL) and 3,30,5-triiod-L-thyronin (0.136 mg/mL). The skin of the seedlings were cut off. The explants, consisting of two cot- fragments were incubated in a humidified atmosphere with CO2 yledons and a hypocotyl, were than cultivated with the tested (5%; v/v) at 37 C. The medium was changed weekly until the fi- cytokinin in a dark room. After 48 h incubation the concentration of broblasts reached confluence. After 2e3 weeks cells were M. Honig€ et al. / European Journal of Medicinal Chemistry 150 (2018) 946e957 955 trypsinized and transferred into 75 cm2 cultivation flasks. Fibro- NHDF/HaCaT. After 60 min incubation, medium with test com- blasts were used between the 2nd and 4th passage. Spontaneously pound was discarded, cells were washed two-times with PBS and immortalized aneuploid human keratinocyte cell line (HaCaT) was PBS-G was applied. To study UVA photoprotection, a plate was bought from CLS (Eppeheim, Germany). Both types of cells were exposed to a cytotoxic dose of UVA radiation (7.5 J/cm2 for NHDF grown in cultivation medium consisted of DMEM supplemented and 10 J/cm2 for HaCaT) using a solar simulator SOL 500 (Dr. Hoenle with fetal calf serum (10%, v/v), penicillin (100 mg/mL) and strep- Technology, Germany) equipped with a H1 filter transmitting tomycin (100 U/mL). For all experiments NHDF and HaCaT were wavelengths of 320e400 nm. To study UVB photoprotection, a seeded onto 96-well plates at a density of 0.8 105 cells/mL or plate was exposed to a cytotoxic dose of UVB radiation (150 mJ/ 1.6 105 cells/mL (0.2 mL per well), respectively. cm2) using the solar simulator equipped with a H2 filter trans- Phototoxic potential of test compounds was determined using mitting wavelengths of 295e320 nm. Intensity of UVA or UVB ra- the validated in vitro NRU phototoxicity test as described by diation was evaluated before each irradiation by UVA- or UVB- Spielmann et al. with some modification [41]. Test substances meter (Dr. Hoenle Technology, Germany). Control (non-irradiated) included compounds 1, 5, 6, 9 and 10. Compounds were dissolved in plates were for the period of irradiation incubated in dark. After DMSO and then diluted in serum free medium (DMEM supple- UVA or UVB exposure, PBS-G was discarded and serum free me- mented with penicillin (100 mg/mL) and streptomycin (100 U/mL)). dium was applied. After 24 h (37 C, 5% CO2) cell damage was At 24 h after seeding, the cultivation medium was changed to evaluated by NR incorporation into viable cells as described above. serum free medium containing test compound or DMSO (a negative The experiments were replicated four times using the cells of four control). The final applied concentration range of test compounds donors to minimize the extent of biological variability of donor was 3.9e125 mM. DMSO was used in the concentration of 0.5%, v/v). cells. The photoprotective effect was evaluated by comparison of In parallel with test compounds, CPZ a known phototoxic com- experimental data (absorbance) of test compounds with a positive pound, was used as a positive control in all experiments. CPZ was control and a negative control (according to the following equation: used in the range of 3.9e500 mM for non-irradiated HaCaT and

0.39e100 mM for NHDF and irradiated HaCaT. The test compounds As Anc Protection ð%Þ¼100 ,100 (six replicate wells per concentration) were in parallel applied on Apc Anc two 96-well plates with NHDF/HaCaT. After 60 min incubation with test compounds medium was discarded, cells were washed two- As … absorbance of sample (cells pre-incubated with test com- times with phosphate buffered saline (PBS) and then PBS supple- pounds in serum free medium and irradiated)Anc … absorbance of mented with glucose (PBS-G, 1 mg/mL) was applied. Randomly, one negative control (cells pre-incubated with s DMSO in serum free plate was then exposed to a non-cytotoxic dose of UVA radiation medium and non-irradiated ¼ incubated in dark)Apc … absorbance (5.0 J/cm2 for NHDF and 7.5 J/cm2 for HaCaT)) using a solar simu- of positive control (cells pre-incubated with s DMSO in serum free lator SOL 500 (Dr. Hoenle Technology, Germany) equipped with a medium and irradiated) H1 filter transmitting wavelengths of 320e400 nm. Intensity of UVA radiation was evaluated before each irradiation by an UVA- 4.7. Oxidative stress bioassays in Caenorhabditis elegans meter (Dr. Hoenle Technology, Germany). The second (non-irradiated) plate was for the period of irradiation incubated in dark. After Wild-type (N2) strain and BA17 (fem-1) strain obtained from UVA exposure, PBS-G was discarded and serum free medium was Caenorhabditis Genetic Center were used in the experiments. applied. After 24 h (37 C, 5% CO2) cell damage was evaluated by NR Buffers and media used in this assay were prepared according to incorporation into viable cells. Medium was discarded and NR so- literature [42]. Worms were maintained at 20 C on NGM plates lution (0.03%, w/v, PBS) was applied. After 60 min NR solution was seeded with Escherichia coli OP50 which was used as a food source. discarded, cells were fixed with a mixture of formaldehyde (0.5%, v/ Age synchronized L4 larvae obtained by hypochlorite treatment v) and CaCl2 (1%, m/v) in ratio 1:1 and then NR was dissolved in were washed from NGM plates with S-complete medium and methanol (50%, v/v) with acetic acid (1%, v/v). After 5 min of number of worms in 10 drops of 10 ml volume was counted. Bac- intensive shaking absorbance was measured at 540 nm. Experi- terial suspension and medium was added to the worm suspension ments were performed in four independent repetitions with use of to achieve 110 worms/mL with 6 mg/mL E. coli OP50 as a food cells from four donors to minimize individual sensitivity of donor source. In experiments with N2 strain, suspension was supple- cells. Phototoxic effect was evaluated as % of viability of control cells mented with 25 mg/mL of 20-deoxy-5-fluorouridine (FUDR) to pre- that was calculated from experimental data (absorbance) according vent reproduction of worms. Age synchronized BA17 worms were to the following equation: cultivated in 25 C which lead to feminisation of the population and addition of FUDR was not necessary. Bacterial suspension in S- ðA A Þ complete medium was prepared thusly: overnight culture of E. coli Viability ð% of controlÞ¼ S B ,100 ðAC ABÞ OP50 in LB medium was centrifuged and bacterium pellets were washed twice with sterile distilled water. After removing the excess AS … absorbance of sample (cells pre-incubated with test com- of water, the pellets were weighed and re-suspended in S-complete pound in serum free medium and irradiated)AC … absorbance of medium. The test compounds were added three days prior to the control (cells pre-incubated with DMSO in serum free medium and exposure to 500 mM of 5-hydroxy-1,4-naphthalenedione (juglone) irradiated)AB … absorbance of background (extraction solution) that is routinely used as a ROS-generating agent in stress assays Photoprotective effects were tested in compounds 1, 5, 6, 9 and with C. elegans [35]. Vehicle (DMSO) was used as a negative control. 10. Compounds were dissolved in DMSO and then diluted in serum The worms were counted under the inverted microscope and the free medium). At 24 h after seeding, cultivation medium was animals that failed to respond to the light stimulus were scored as changed to serum free medium containing test compound or DMSO dead. The protective effect of the compounds was evaluated after (a negative control). The final applied concentration range of test 4 h in the initial screening on populations of approximately 110 compounds was 3.9e31.3 mM. DMSO was used in the concentration worms per experimental condition. The experiment was repeated 3 of 0.5%, v/v. RA, a known photoprotective compound was used for times. Statistical significance of the difference between proportions comparison of test compounds photoprotective effectiveness. Each of living worms was determined using z-test. In the follow-up ex- test compound was in parallel applied onto two 96-well plates with periments on a larger populations worms (approximately 250 or 956 M. Honig€ et al. / European Journal of Medicinal Chemistry 150 (2018) 946e957

190 worms on average), the activity of selected promising com- [7] R. Zhang, L.S. David, Cytokinin biochemistry in relation to leaf senescence. III. pounds was evaluated hourly for 12e14 h. Log-rank test was used The senescence-retarding activity and metabolism of 9-substituted 6- benzylaminopurines in soybean leaves, J. Plant Growth Regul. 8 (1989) to determine the statistical significance of differences in the sur- 181e197. vival distributions. P-values were adjusted for multiple testing us- [8] V. Mik, L. Szücova, M. Smehilova, M. Zatloukal, K. Dolezal, J. Nisler, J. Grúz, < P. Galuszka, M. Strnad, L. Spíchal, N9-substituted derivatives of kinetin: ing Bonferroni method. Comparisons with p-values 0.05 were e fi effective anti-senescence agents, Phytochemistry 72 (2011) 821 831, https:// considered as statistically signi cant. OASIS 2 was used for the data doi.org/10.1016/j.phytochem.2011.02.002. analysis [43]. [9] R.S. Dhindsa, P.L. Plumb-Dhindsa, D.M. Reid, Leaf senescence and lipid peroxidation: effects of some phytohormones, and scavengers of free radicals and singlet oxygen, Physiol. Plantarum 56 (1982) 453e457, https://doi.org/ 4.8. Oxygen radical absorbance capacity (ORAC) 10.1111/j.1399-3054.1982.tb04539.x. [10] B. Griffaut, R. Bos, J.C. Maurizis, J.C. Madelmont, G. Ledoigt, Cytotoxic effects of The oxygen radical absorbance capacity (ORAC) was determined kinetin riboside on mouse, human and plant tumour cells, Int. J. Biol. Mac- romol. 34 (2004) 271e275, https://doi.org/10.1016/j.ijbiomac.2004.06.004. as described previously [44]. Briefly, fluorescein (100 mL, 500 mM) [11] J. Voller, T. Beres, M. Zatloukal, P.A. Kaminski, P. Niemann, K. Dolezal, and solutions of tested compounds diluted in phosphate buffer P. Dzub ak, M. Hajdúch, M. Strnad, The natural cytokinin 2OH3MeOBAR induces cell death by a mechanism that is different from that of the “classical” were added into each working well of a 96-well microplate. The e 0 cytokinin ribosides, Phytochemistry 136 (2017) 156 164, https://doi.org/ reaction was started by the addition of 2,2 -azobis (2-amidino- 10.1016/j.phytochem.2017.01.004. propane) dihydrochloride (AAPH; 25 mL, 250 mM) and the decrease [12] J. Veselý, L. Havlicek, M. Strnad, J.J. Blow, A. Donella-Deana, L. Pinna, in fluorescence (ex. 485 nm, em. 510 nm) was recorded every 3 min D.S. Letham, J.y Kato, L. Detivaud, S. Leclerc, L. Meijer, Inhibition of cyclin- fi dependent kinases by purine analogues, Eur. J. Biochem. 224 (1994) over 90 min by using In nite M200 Pro (Tecan, Switzerland). The 771e786, https://doi.org/10.1111/j.1432-1033.1994.00771.x. net area under the curve was used to calculate antioxidant capacity [13] W.F. De Azevedo, S. Leclerc, L. Meijer, L. Havlicek, M. Strnad, S.H. Kim, Inhi- which was compared to Trolox and expressed as Trolox equivalents bition of cyclin-dependent kinases by purine analogues: crystal structure of human cdk2 complexed with roscovitine, Eur. J. Biochem. 243 (1997) (TE). Thus, a value higher than 1 means that the tested compound 518e526 doi:9030780. was more active than Trolox on equimolar basis. [14] A. Olsen, G.E. Siboska, B.F.C. Clark, S.I.S. Rattan, N6-Furfuryladenine, kinetin, protects against fenton reaction-mediated oxidative damage to DNA, Bio- chem. Biophys. Res. Commun. 265 (1999) 499e502, https://doi.org/10.1006/ Acknowledgements bbrc.1999.1669. [15] S.P. Sharma, P. Kaur, S.I.S. Rattan, Plant growth hormone kinetin delays ageing, fl This work was supported by the Ministry of Education, Youth prolongs the lifespan and slows down development of the fruit y Zaprionus paravittiger, Biochem. Biophys. Res. Commun. 216 (1995) 1067e1071. http:// and Sports, Czech Republic particularly by the grant No. LO1204 search.ebscohost.com/login.aspx?direct¼true&db¼edsagr&AN¼edsagr. from the National Program of Sustainability I and Agricultural US201301513764&site¼eds-live&authtype¼shib&custid¼s7108593. Research, by TACR GAMA “Effective Transfer of Palacký University [16] S.P. Sharma, J. Kaur, S.I.S. Rattan, Increased longevity of kinetin-fed Zaprionus fl ” fruit ies is accompanied by their reduced fecundity and enhanced catalase Olomouc Results into Praxis TG01010080 and by the Institutional activity, Biochem. Mol. Biol. Int. 51 (1997) 869e875. http://search.ebscohost. Support of Palacký University in Olomouc (RVO 61989592) as well com/login.aspx?direct¼true&db¼edsagr&AN¼edsagr. as by the Internal Grant Agency of Palacký University US201302864121&site¼eds-live&authtype¼shib&custid¼s7108593. [17] M. Li, W. Ouyang, J. Li, L. Si, X. Li, J. Guo, H. Li, Effects of kinetin on thymus and (IGA_PrF_2018_023). We would like to thank Jarmila Balonova, Olga e immune function of aging rats, Pak. Vet. J. 36 (2016) 356 362. https:// Hustakov a and Miroslava Subova for their skillful technical assis- shibboleth.ebscohost.com/Shibboleth.sso/Login?providerId¼https://idp.upol. tance and Mgr. Tomas Pospísil Ph.D. for his assistance with mea- cz/idp/shibboleth&target¼http://search.ebscohost.com/login.aspx? 1 " direct¼true&db¼a9h&AN¼116996096&lang¼cs&site¼eds-live. surement of H NMR spectral data. C. elegans strain N2 was [18] A. Jabłonska-Trypu c, M. Matejczyk, R. Czerpak, N6-benzyladenine and kinetin provided by the CGC, which is funded by NIH Office of Research influence antioxidative stress parameters in human skin fibroblasts, Mol. Cell. Infrastructure Programs (P40 OD010440)." Biochem. 413 (2016) 97e107, https://doi.org/10.1007/s11010-015-2642-5. [19] E.M. Othman, M. Naseem, E. Awad, T. Dandekar, H. Stopper, The plant hormone cytokinin confers protection against oxidative stress in mammalian Appendix A. Supplementary data cells, PLoS One 11 (2016) 1e14, https://doi.org/10.1371/ journal.pone.0168386. [20] A. Brizzolari, C. Marinello, M. Carini, E. Santaniello, P.A. Biondi, Evaluation of Supplementary data related to this article can be found at the antioxidant activity and capacity of some natural N6-substituted adenine https://doi.org/10.1016/j.ejmech.2018.03.043. derivatives (cytokinins) by fluorimetric and spectrophotometric assays, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 1019 (2016) 164e168, https://doi.org/10.1016/j.jchromb.2015.12.047. References [21] L. Szücova, M. Zatloukal, L. Spíchal, L. Frohlich,€ K. Dolezal, M. Strnad, F. Massino, 6,9-Disubstituted Purine Derivatives and Their Use for Treating [1] P.J. Davies, Plant hormones, springer Netherlands, Dordrecht (2010), https:// Skin, 2008. WO2008008770A3 and US2008009508. doi.org/10.1007/978-1-4020-2686-7. [22] S.I.S. Rattan, B.F.C. Clark, Kinetin delays the onset of aging characteristics in [2] C.O. Miller, F. Skoog, F.S. Okomura, M.H. Saltza, F.M. Strong, Isolation, structure human fibroblasts, Biochem. Biophys. Res. Commun. 201 (1994) 665e672. and synthesis of kinetin, a substrance promoting cell division, J. Am. Chem. [23] S.I.S. Rattan, N 6 -furfuryladenine (kinetin) as a potential anti-aging molecule, Soc. 78 (1956) 1375e1380, https://doi.org/10.1021/ja01588a032. J. Anti Aging Med. 5 (2002) 113e116, https://doi.org/10.1089/ [3] L. Ge, J.W.H. Yong, N.K. Goh, L.S. Chia, S.N. Tan, E.S. Ong, Identification of 109454502317629336. kinetin and kinetin riboside in coconut (Cocos nucifera L.) water using a [24] J.L. McCullough, R.L. Garcia, B. Reece, A clinical study of topical Pyratine 6 for combined approach of liquid chromatography-tandem mass spectrometry, improving the appearance of photodamaged skin, J. Drugs Dermatol 7 (2008) high performance liquid chromatography and capillary electrophoresis, 131e135. http://www.ncbi.nlm.nih.gov/pubmed/18335649. J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 829 (2005) 26e34, https:// [25] J. Vicanova, C. Bouez, S. Lacroix, L. Lindmark, O. Damour, Epidermal and doi.org/10.1016/j.jchromb.2005.09.026. dermal characteristics in skin equivalent after systemic and topical applica- [4] J. Barciszewski, F. Massino, B.F.C. Clark, Kinetin-A multiactive molecule, Int. J. tion of skin care ingredients, Ann. N. Y. Acad. Sci. 1067 (2006) 337e342, Biol. Macromol. 40 (2007) 182e192, https://doi.org/10.1016/ https://doi.org/10.1196/annals.1354.046. j.ijbiomac.2006.06.024. [26] P.M.B.G. Maia Campos, F.B. de Camargo Júnior, J.P. de Andrade, L.R. Gaspar, [5] K. Dolezal, I. Popa, E. Hauserova, L. Spíchal, K. Chakrabarty, O. Novak, Efficacy of cosmetic formulations containing dispersion of liposome with V. Krystof, J. Voller, J. Holub, M. Strnad, Preparation, biological activity and magnesium ascorbyl phosphate, alpha-lipoic acid and kinetin, Photochem. endogenous occurrence of N6-benzyladenosines, Bioorg. Med. Chem. 15 Photobiol. 88 (2012) 748e752, https://doi.org/10.1111/j.1751- (2007) 3737e3747, https://doi.org/10.1016/j.bmc.2007.03.038. 1097.2012.01086.x. [6] L. Szücova, L. Spíchal, K. Dolezal, M. Zatloukal, J. Greplova, P. Galuszka, [27] T. Kaletta, M.O. Hengartner, Finding function in novel targets: C. elegans as a V. Krystof, J. Voller, I. Popa, F.J. Massino, J.E. Jørgensen, M. Strnad, Synthesis, model organism, Nat. Rev. Drug Discov. 5 (2006) 387e399, https://doi.org/ characterization and biological activity of ring-substituted 6-benzylamino-9- 10.1038/nrd2031. tetrahydropyran-2-yl and 9-tetrahydrofuran-2-ylpurine derivatives, Bioorg. [28] L.P. O'Reilly, C.J. Luke, D.H. Perlmutter, G.A. Silverman, S.C. PakC, Elegans in Med. Chem. 17 (2009) 1938e1947, https://doi.org/10.1016/ high-throughput drug discovery, Adv. Drug Deliv. Rev. 69e70 (2014) j.bmc.2009.01.041. 247e253, https://doi.org/10.1016/j.addr.2013.12.001. M. Honig€ et al. / European Journal of Medicinal Chemistry 150 (2018) 946e957 957

[29] D. De Roulet, R. Devita, Compositions and Methods Using the Same for https://doi.org/10.3109/09553002.2010.501842. Treatment of Neurodegenerative and Mitochondrial Disease, 2015. [39] J. Holub, J. Hanus, D.E. Hanke, M. Strnad, Biological activity of cytokinins WO2015123365. derived from Ortho- and Meta-hydroxybenzyladenine, Plant Growth Regul. [30] M. Bullock, J. Hand, R. Hall, Method of Prparing Substituted Purines, 1962. 26 (1998) 109e115, https://doi.org/10.1023/A:1006192619432. US3041340A. [40] J. Nisler, D. Kopecný, R. Koncitíkova, M. Zatloukal, V. Bazgier, K. Berka, [31] T.W. Greene, P.G.M. Wuts, Protective Groups in Organic Synthesis, John Wiley D. Zalabak, P. Briozzo, M. Strnad, L. Spíchal, Novel thidiazuron-derived in- & Sons, Inc., New York, USA, 1999, https://doi.org/10.1002/0471220574. hibitors of cytokinin oxidase/dehydrogenase, Plant Mol. Biol. 92 (2016) [32] F. Skoog, H.Q. Hamzi, A.M. Szweykowska, N.J. Leonard, K.L. Carraway, T. Fujii, 235e248, https://doi.org/10.1007/s11103-016-0509-0. J.P. Helgeson, R.N. Loeppky, Cytokinins: structure/activity relationships, [41] H. Spielmann, M. Balls, J. Dupuis, W.J. Pape, G. Pechovitch, O. De Silva, Phytochemistry 6 (1967) 1169e1192, https://doi.org/10.1016/S0031- H.G. Holzhütter, R. Clothier, P. Desolle, F. Gerberick, M. Liebsch, W.W. Lovell, 9422(00)86080-X. T. Maurer, U. Pfannenbecker, J.M. Potthast, M. Csato, D. Sladowski, W. Steiling, [33] S. Matsubara, Structure-activity relationships of cytokinins, Phytochemistry P. Brantom, The international EU/COLIPA in vitro phototoxicity validation 19 (1980) 2239e2253, https://doi.org/10.1016/S0031-9422(00)91006-9. study: results of phase II (blind trial). Part 1: the 3T3 NRU phototoxicity test, [34] S. Matsubara, Structure-activity relationships of cytokinins, CRC Crit. Rev. Toxicol. Vitro 12 (1998) 305e327, https://doi.org/10.1016/S0887-2333(98) Plant Sci. 9 (1990) 17e57, https://doi.org/10.1080/07352689009382281. 00006-X. [35] E. De Castro, S.H. De Castro, T.E. Johnson, Isolation of long-lived mutants in [42] G.M. Solis, M. Petrascheck, Measuring Caenorhabditis elegans life span in 96 Caenorhabditis elegans using selection for resistance to juglone, Free Radic. well microtiter plates, JoVE (2011), https://doi.org/10.3791/2496. Biol. Med. 37 (2004) 139e145, https://doi.org/10.1016/ [43] S.K. Han, D. Lee, H. Lee, D. Kim, H.G. Son, J.-S. Yang, S.-J.V. Lee, S. Kim, OASIS 2: j.freeradbiomed.2004.04.021. online application for survival analysis 2 with features for the analysis of [36] D. Kligman, Cosmeceuticals, dermatologic asp, Cosmet 18 (2000) 609e615. maximal lifespan and healthspan in aging research, Oncotarget 7 (2016), [37] M. Sanchez-Campillo, J.A. Gabaldon, J. Castillo, O. Benavente-García, M.J. Del https://doi.org/10.18632/oncotarget.11269, 59147e56152. Bano,~ M. Alcaraz, V. Vicente, N. Alvarez, J.A. Lozano, Rosmarinic acid, a photo- [44] B. Ou, M. Hampsch-Woodill, R.L. Prior, Development and validation of an protective agent against UV and other ionizing radiations, Food Chem. Toxicol. improved oxygen radical absorbance capacity assay using ﬂuorescein as the 47 (2009) 386e392, https://doi.org/10.1016/j.fct.2008.11.026. ﬂuorescent probe, J. Agric. Food Chem. 49 (2001) 4619e4626, https://doi.org/ [38] A. Svobodova, J. Vostalov a, Solar radiation induced skin damage: review of 10.1021/jf010586o. protective and preventive options, Int. J. Radiat. Biol. 86 (2010) 999e1030, 7.4 Supplement 4

Voller, J., Maková, B., Kadlecová, A., Gonzalez, G., & Strnad, M. (2017). Plant hormone cytokinins for modulating human aging and age-related diseases. In Hormones in Ageing and Longevity (pp. 311-335). Springer, Cham.

82 Chapter 14 Plant Hormone Cytokinins for Modulating Human Aging and Age-Related Diseases

Jiří Voller, Barbara Maková, Alena Kadlecová, Gabriel Gonzalez and Miroslav Strnad

Abstract Cytokinins are phytohormones that regulate plant growth, development and senescence. Experiments both in vitro and in vivo demonstrate that they can also have diverse effects on animal cells and tissues. Particularly interesting is their ability to protect cells against various forms of stress and prevent some detrimental effects of cell aging. For example, human skin ﬁbroblasts cultured in the presence of kinetin or trans-zeatin retain some characteristics of cells of lower passage. Kinetin is even able to increase the lifespan of invertebrates. In this chapter, we review protective effects of cytokinins in animals at molecular, cellular, tissue and organismal levels. We also discuss potential application of cytokinins for the treatment of age-related diseases, including neurodegenerations, inﬂammatory diseases and disorders caused by aberrant cell proliferation.

Keywords Cytokinin Á Kinetin Á Kinetin riboside Á Zeatin Á Benzyl adenine Á Topolin Á Skin Á Anti-aging Á Anti-inﬂammatory activity Á Neuro protection

List of abbreviations ADK Adenosine kinase A2ARA2A adenosine receptor A3RA3R adenosine receptor BA N6-benzyladenine BAR N6-benzyladenosine ENT Equilibrative nucleoside transporter tZ trans-zeatin tZR trans-zeatin riboside iP N6-isopentenyladenine iPR N6-isopentenyladenosine ipt Isopentenyltransferase

J. Voller (&) Á B. Maková Á A. Kadlecová Á G. Gonzalez Á M. Strnad Laboratory of Growth Regulators, Centre of the Region Haná for Biotechnological and Agricultural Research, Institute of Experimental Botany ASCR, Palacký University, Šlechtitelů 27, 78371 Olomouc, Czech Republic e-mail: [email protected]

K Kinetin KR Kinetin riboside KRTP Kinetin riboside-5′-triphosphate oTR ortho-topolin riboside pTR para-topolin riboside ROS Reactive oxygen species

14.1 Introduction

Cytokinins are phytohormones identified originally as substances that promote plant cell division in the presence of another phytohormone, auxin (Skoog et al. 1965). They play important roles in the development and growth of both root and shoot systems. Processes regulated by cytokinins include water and nutrient mobilization, apical dominance, branching, flowering, breaking of bud dormancy and seed germination (Werner and Schmülling 2009). Cytokinins also delay leaf senescence—they prevent the degradation of chlorophyll and outflow of nutrients from the leaf. A good demonstration of this activity occurs in “evergreen” transgenic tobacco plants with the gene encoding ipt (isopentenyltransferase), the protein responsible for cytokinin biosynthesis under the control of a senescence-specific promoter (Gan and Amasino 1995). Knowledge that cytokinins play key roles in the regulation of plant growth and development has stimulated studies into their potential utility for treating human diseases, especially those involving dysfunctional cell proliferation and/or differentiation. Cytokinin ribosides were even evaluated in patients with diverse malignancies as early as the 1970s (Mittelman et al. 1975). Later on, cytokinins inspired development of inhibitors of cyclin-dependent kinases olomoucine, bohemine, roscovitine and their analogues (Veselý et al. 1994; Havlíček et al. 1997; Vermeulen et al. 2002; Kryštof et al. 2002). The ability of cytokinins to prevent processes associated with plant senescence has attracted research focused on their ability to ameliorate aging traits in animals, including humans. In this article, we review the effects of cytokinins in animals at molecular, cellular, tissue and organismal levels with an emphasis on anti-aging and cytoprotective activity. We also discuss potential application of cytokinins in the treatment of inflammation and neurodegeneration as relevant to various age-related diseases. Finally, we include a section about the antiproliferative activity of cytokinins—they are cytotoxic for malignant cells of diverse histopathological origin but are also able to induce differentiation of some leukemia cells and keratinocytes. 14 Plant Hormone Cytokinins for Modulating Human Aging … 313

14.2 Chemical Structure and Occurrence of Cytokinins

Regarding their chemical structure, plant cytokinins are adenine derivatives substituted at the N6-position with either an isoprenoid or aromatic side chain. Synthetic compounds with cytokinin activity can have other scaffolds, phenylureas being the best known example. Isoprenoid cytokinins include cis- and trans-zeatin (tZ) and their analogues with a saturated side chain (dihydrozeatin) or without an hydroxyl group (N6-isopentenyladenine, iP). Whereas isoprenoid cytokinins are present in all plants, aromatic cytokinins with N6-benzyl substituents have only been found in certain taxa (Horgan et al. 1975; Strnad 1997). Besides N6-benzyladenine (BA), its hydroxylated derivatives, i.e., topolins, have been described. Kinetin (K), the first identified cytokinin, has an N6-furfuryl side chain. K was first recognized as the substance responsible for the cytokinin activity of autoclaved herring sperm, which was attributed to thermal DNA damage. Later it was reported to occur naturally in plant material (Ge et al. 2004), as well as human cells and urine (Barciszewski et al. 1996, 2000). However, its natural occurrence is probably very rare and rather mysterious. We have not even been able to unequivocally identify this compound in diverse plant material using sophisticated tandem mass spectrometry analysis. Barciszewski et al. (1997) proposed that endogenous K is a result of oxidative DNA damage. Both isoprenoid and aromatic cytokinins occur as free bases, ribosides, ribotides, N-glucosides and amino acid conjugates. Moreover cytokinins with hydroxylated isoprenoid side chains can also form O-glycosides. N6-substituted adenine derivatives also occur in certain tRNA species of all organisms with the exception of Archea. For example, in mammals, an N6-isopentenyladenosine (iPR) moiety forms part of tRNA[Ser]Sec. It facilitates codon-anticodon interactions and contributes to the efficiency of selenoprotein synthesis. Adenines and adenosines with N6-substitution may be released into the cytosol, and subsequently into body fluids, as a result of tRNA turnover. iPR was detected in human urine many years ago (Chheda and Mittelman 1972) (Fig. 14.1).

Fig. 14.1 Natural cytokinin bases studied as modulators of processes related to aging. From left to right—kinetin, trans-zeatin, N6-isopentenyladenine, N6-benzyladenine, para-topolin 314 J. Voller et al.

14.3 Cytokinin Signaling in Plants

The cytokinin signal in plants is perceived by a His-Asp phosphorelay similar to the two-component systems of bacteria. After recognition of the cytokinin ligand by the extracellular domain of the transmembrane cytokinin receptor (AHK2, AHK3, or AHK4 in Arabidopsis thaliana), the intracellular portion of the receptor phos- phorylates histidine phosphotransfer proteins (AHPs). These transmit the signal to nuclear response regulators (ARRs), which can activate or repress transcription of the response genes. Anti-senescence activity of cytokinins is mediated through the activation of AHK3, the type-B response regulator ARR2 and cytokinin response factor CRF6. Increased cell-wall invertase activity in response to cytokinins is both necessary and sufficient for the inhibition of senescence (Zwack and Rashotte 2013). Although differences in the substrate specificity between individual cytokinin receptors exist, cytokinin bases are consistently the most active cytokinin form in both receptor assays and cytokinin biotests (Mok and Mok 2001;Spíchal et al. 2004). The intensity and duration of the signaling is dependent on the receptor and response regulator composition of the given cell/tissue and the availability of individual cytokinins. The rate-limiting step in cytokinin biosynthesis is catalyzed by isopentenyltransferases (IPTs), which synthesize either free cytokinin nucleotides (adenosine phosphate-IPTs) or modify adenosine in tRNA (tRNA-IPTs). Conversion of cytokinin 5′-monophosphates into their respective free bases is catalyzed by phosphoribohydrolase encoded by the gene LONELY GUY (LOG) (Kurakawa et al. 2007). An alternative pathway where dephosphorylation of riboside-5′-monophosphates precedes the cleavage of the glycoside bond also exists (Chen and Kristopeit 1981), but the genes responsible have not yet been characterized. Cytokinins are degraded by cytokinin oxidase/dehydrogenases (CKXs), which catalyze removal of the side chain. The cytokinin signal is also attenuated by conversion of free bases into less active (ribosides, ribotides) or inactive forms (glucosides, conjugates with alanine). With the exception of N7- and N9-glucosides, cytokinin conjugates can be converted back into free bases and are seen as transport/storage cytokinin forms. The uptake and efflux of cytokinins by cells is facilitated by members of the purine permease family (PUPs) of transmembrane channels (Gillissen et al. 2000) and by equilibrative nucleoside transporters (ENTs) (Hirose et al. 2008). Cytokinins are present in both phloem and xylem fluid and serve as both acropetal and basipetal messengers (Kudo et al. 2010). The first acropetal transporter was described recently (Zhang et al. 2014). 14 Plant Hormone Cytokinins for Modulating Human Aging … 315

14.4 Effects of Cytokinins in Animal Systems

14.4.1 Cytoprotective and Anti-aging Activity of Cytokinins

Interest in the anti-aging activity of cytokinins started in 1994 when Rattan and Clark discovered positive effects of K on several characteristics related to aging in human skin fibroblasts during serial passage in vitro (Rattan and Clark 1994). The size and morphology of fibroblasts passaged in the presence of K resembled those of the cells at lower passage numbers. Treatment with K decreased the number of actin stress fibers and autofluorescence due to accumulated lipofuscin was also less intense. Similar effects of tZ on in vitro aging of a fibroblast population were reported more than 10 years later (Rattan and Sodagam 2005). Optimal anti-aging effects were observed at 80 lM concentration for both compounds. It is important to note that these long-term cultivation experiments were enabled by the remarkably low toxicity of the cytokinins tested. Beneficial effects of several other cytokinin bases on various parameters relevant for aging amelioration and/or age-related disease therapy were reported in the years following the original discovery, and we discuss them below. Active compounds include para-topolin, iP and a K derivative 6-furfurylamino-9-(tetrahydropyran-2-yl)-9H-purine (Pyratine-6, PRK-124) (Walla et al. 2010). K, tZ and Pyratine are the principal ingredients of several marketed cosmeceuticals. Notably, K effects related to aging are not limited to cells and tissues as dietary K has been shown to increase the life span of Zaprionus paravittiger fruit flies. K prolonged the larval and pupal stages but also reduced the age-specific death rates throughout the adult lifespan (Sharma et al. 1995). The effect was accompanied by enhanced catalase activity (Sharma et al. 1997). We also recently discovered that K and several other cytokinin derivatives are able to increase the lifespan of Caenorhabditis elegans (our unpublished data).

14.4.2 Anti-oxidant Activity of Cytokinins

Since the discovery of anti-aging activity of K in human fibroblasts, research has primarily focused on its ability to protect macromolecules and cells against oxidative damage and stress. Barciszewski et al. (1997) hypothesized that endogenous K may arise as a consequence of oxidative damage of DNA, thus creating protection near to the site of damage. The proposed mechanism of K formation assumes that hydroxy radical attack at the 5′ carbon of the deoxyribose residue yields furfural. This aldehyde reacts with the amino group of adenine and, after intramolecular rearrangement, the resulting Schiff base is reduced into K (Barciszewski et al. 1997). Using 8-oxo-2′-deoxyguanosine (8-oxo-dG) as a marker for oxidative damage of DNA, Olsen et al. (1999) showed that K significantly protects DNA against reactive 316 J. Voller et al. oxygen species (ROS) generated by the Fenton reaction in vitro. The effect was dose dependent, with a maximum of about 50% protection observed at 100 lMK. K has also been shown to protect proteins against oxidative/glycoxidative damage more efficiently than adenine in several experimental systems in vitro (Olsen et al. 1999). Besides decreasing protein carbonylation in an iron/ascorbate system, it also prevented formation of advanced glycation end-product pentosidine and aggregation after incubation of proteins with sugars. Active K concentrations were in the range 50–200 lM (Verbeke et al. 2000). Recently, the antioxidant activities of K, BA, para-topolin and iP were evaluated by fluorimetric and spectrophotometric assays (Brizzolari et al. 2016). With the exception of BA, all the compounds showed significant activity in the oxygen radical absorbance capacity (ORAC) assay at 2.5 and 5 lM concentrations. In the Trolox equivalence antioxidant capacity (TEAC) assay, only para-topolin (0.5–5 lM) was active, probably due to the presence of a phenolic hydroxyl. All the compounds were able to react with hydroxyl radicals generated in the 2-deoxyribose degradation assay. A somewhat higher activity of iP was ascribed to the presence of the double bond in the N6-side chain. Using electron spin resonance, Hsiao et al. (2003) showed that short pre- treatment with K at concentrations of 70 and 150 lM effectively inhibited hydroxyl radical formation in collagen-activated platelets. Direct comparison of K with a group of other established antioxidants, comprising L-ascorbic acid, DL-alpha-tocopherol, DL-alpha lipoic acid, ubiquinone and idebenone, has been carried out (McDaniel et al. 2005). K quenched radicals generated through the photochemical excitation of water molecules effectively only at a concentration of 1 lM. The effective concentrations of the other compounds, with the exception of DL-alpha lipoic acid, were one or two orders of magnitude lower. K (100 lM) completely prevented oxidation of low density lipoproteins by Cu2+. The other compounds decreased the production of lipid hydroxyperoxides in equimolar concentrations less efficiently (idebenone: 80%, DL-alpha-tocopherol: 50%, ubiquinone: 26%, DL-alpha lipoic acid: 22%, L-ascorbic acid: 15%). However, this exceptional activity of K was not observed in the microsome oxidation (NADPH/ADP/Fe3+) assay, which is considered to be a more realistic model of cell membrane peroxidation. Whereas the activity of K was comparable with that of ubiquinone and L-ascorbic acid (reduction of malonyldialdehyde equivalents by 24–27%), the other test compounds decreased MDA production by 47–55%. Cytokinins may also form complexes with ions of metals with the ability to quench ROS. Superoxide dismutase-like activity of Cu2+ complexes of K and BA have been reported (Goldstein and Czapski 1991). The ability of Cu2+ complexes of BA derivatives to protect against oxidative damage in vivo (aloxan induced diabetes) has been observed (Štarha et al. 2009). Besides their direct anti-oxidant activities, cytokinins also activate cellular anti-oxidant defense mechanisms. tZ has been shown to induce hydrogen peroxide decomposing enzymes in human skin fibroblasts. The treatment was also able to protect both proliferating and senescent cells against the hydrogen peroxide induced cell death (Rattan and Sodagam 2005). Induction of antioxidant enzymes may also contribute to the lifespan extension observed in flies after K treatment mentioned 14 Plant Hormone Cytokinins for Modulating Human Aging … 317 above (Sharma et al. 1997) because it enhances the catalase activity in this organism. K also has been demonstrated to exhibit protective effects in the D- galactose model of glycoxidative stress. In cultured rat astrocytes, K partially reversed a decrease in the activities of glutathione peroxidase and superoxide dismutase induced by D-galactose treatment. K treatment also decreased malonyldialdehyde concentration in the cell membranes and increased cell viability. The active concentrations were in the range 50–100 µM. In addition, K was shown to have beneficial effects in rats receiving daily subcutaneous injections of D-galactose at 125 mg/kg for 6 weeks. K (10, 20 and 40 mg/kg) was administered by gastric perfusion for the whole period of galactose exposition. However, no information was provided about K concentrations in plasma or brain tissue. A dose of 10 mg/kg was concluded to have the most promising effect because it was able to attenuate the negative effects of D-galactose on malonyldialdehyde concentration and activities of glutathione peroxidase and superoxide dismutase in brain tissue (Liu et al. 2011). Although iPR and N6-benzyladenosine (BAR) have been studied traditionally as anti-cancer agents, they possess cytoprotective activities as well. Dassano et al. (2014) showed that treatment of several cancer cell lines with iPR or BAR induced robust expression of genes regulated by transcription factor Nrf2. Moreover, the transcription of Nrf2 itself was upregulated. These expression changes were not a just a futile response to oxidative stress resulting from drug-induced damage. In fact, the treatment with cytokinin ribosides decreased the basal levels of ROS and increased the resistance of the cells to pro-oxidative insults. Our earlier unpublished observation that not only iPR but also other cytotoxic ribosides, including kinetin riboside (KR), in low micromolar concentrations induce Nrf2-dependent gene expression in multiple cell lines led us to speculate that the cytoprotective activity of K could be mediated by its metabolite KR and/or appropriate ribotides. The conversion of cytokinin bases into their respective riboside 5′-monophosphates by human cells has been reported (Mlejnek and Doležel 2005). However, treatment of several cell lines, including skin fibroblasts and immortalized keratinocytes, with 100 lM K did not have any significant effect on hemeoxygenase expression (unpublished data). Because tissues may differ in expression and/or activity of phos- phoribosyltransferase, and thus in the rate of K conversion into its riboside, studies of a wider panel of cell lines could identify tissues where the proposed mechanism is relevant. Nevertheless, Hertz et al. (2013) recently demonstrated that conversion of K into its ribotides may be indeed important for its cytoprotective activity. Mitoprotective PTEN-induced putative kinase 1 (PINK1, PARK6) is able to use kinetin riboside 5′-triphosphate (KRTP) as a donor of a phosphate group more efficiently than ATP. Because of its unusual behavior, Hertz et al. (2013) designated KRTP as a PINK1 neo-substrate. They showed that K is converted into KRTP within the cells and that treatment with K accelerated Parkin recruitment to depolarized mitochondria and suppressed oxidative stress-induced apoptosis in human-derived neural cells in a PINK1-dependent manner. 318 J. Voller et al.

14.4.3 Cytokinins in the Therapy of Skin Diseases and Cosmetics

Because the anti-aging activity of K, and later tZ, was demonstrated for the first time in skin fibroblasts, multiple studies evaluating the applicability of cytokinin bases in skin protection both in vitro and in vivo have been conducted. Positive effects include protection against UV-induced damage, improved wound healing and aquaporin induction. Cytokinins also modulate melanogenesis and keratinocyte differentiation. Several clinical trials have shown that they can improve multiple traits of photoaging skin and also some symptoms of acne rosacea. McDaniel et al. (2005) compared the UV-protective activity of a group of antioxidants, i.e., K, L-ascorbic acid, DL-alpha-tocopherol, DL-alpha lipoic acid, ubiquinone and idebenone. K was shown to protect primary keratinocytes against UVB (single dose, 200 mJ/cm2); it reduced the number of cells stained positively by immunohistochemistry with an antibody against thymine–dimer from 53 to 34%. Whereas idebenone, L-ascorbic acid and DL-alpha-tocopherol offered similar protection levels (29–35% of positive cells), ubiquinone and DL lipoic acid were not active. Subsequently, the UV-protective activity of the compounds was tested in patients, each compound on five subjects. The compounds were applied as ethanolic solutions (0.5% w/w) on mid-back regions once a day for 2 weeks. Sunburn cells (SBC) induced by a 1.5 Â minimal erythema dose of UVB were quantified in biopsies obtained 20 h later. K reduced the number of SBC by 20%, compared to about 10% for ubiquinone and DL-lipoic. Whereas idebenone and DL- alpha-tocopherol were more active (reduction by 38 and 30%, respectively), ascorbic acid did not have any protective effect. Unfortunately, no UV protective effects of K were observed in pigs. Four days’ application of 0.1% K cream or 0.5% K solution neither ameliorated UV-induced erythema nor reduced the number of SBC in pigs (Tournas et al. 2006) exposed to a one to five times minimal erythema dose of UV irradiation (*5 mW/cm2 of UVB and *40 mW/cm2 of UVA). Cell culture experiments have suggested that tZ could improve skin hydration and wound healing, as well as prevent detrimental effects of photoaging on those processes (Ji et al. 2010). tZ at 40 and 80 lM concentration also induced expression of aquaporin 3 protein (AQP3) in spontaneously immortalized HaCaT keratinocytes. Treatment with tZ was also able to ameliorate a UV-induced decrease in AQP3 concentrations and membrane water permeability to a large extent. Experiments with pharmacological MAPK pathway inhibitors revealed that tZ inhibits UV-induced MEK/ERK activation. tZ has also been reported to promote wound healing in the scratch assay with either irradiated or non-irradiated cell cultures (Ji et al. 2010) and inhibit (tZ at 20–40 lM) UVB-induced MMP-1 expression in skin fibroblasts (Yang et al. 2009). Cytokinin bases have been shown to promote differentiation of keratinocytes. Kat40–200 lM concentration induced growth arrest and changes of several markers of differentiation (keratin K10 and involucrin) in human keratinocytes in cell culture (Berge et al. 2006). The effect was augmented by the presence of Ca2+ 14 Plant Hormone Cytokinins for Modulating Human Aging … 319 ions. Other markers of differentiation (trans-glutaminase) were unchanged, suggesting that K-induced differentiation might be mediated by pathways different from those activated by other differentiation inducing agents. In a subsequent study, Berge et al. (2008) reported that treatment with K improved the sensitivity of aging keratinocytes to the differentiating effects of Ca2+ ions. The authors mentioned unpublished data indicating that the effect was accompanied by induction of Hsp90, Hsp70 and heme-oxygenase-1. They suggested that the effects were mediated by stress-induced hormesis based on this observation. A positive effect of K on filaggrin levels, another marker of keratinocyte differentiation, was observed in an in vitro reconstructed skin equivalent (Vičanová et al. 2006). In contrast to 2D culture, K promoted the growth of keratinocytes, as indicated by an increase in the number of Ki67-positive cells. Both experimental systems mentioned above used human keratinocytes from healthy donors. However, it should be noted that models using keratinocytes from psoriatic lesions might be more appropriate for evaluation of the utility of cytokinins in the therapy of psoriasis. Other dermatologic/cosmetic applications of cytokinins relate to the therapy of pigmentation disorders. Whereas K has been reported to decrease hyperpigmentation in dogs (Kimura and Doi 2004), BA is a stimulator of melanogenesis (Kim et al. 2009). At concentrations of 50 and 100 lM, it stimulated melanogenesis in B16 mouse melanoma cells through protein kinase A mediated induction of microphthalmia- associated transcription factor (MITF), tyrosinase and tyrosinase-related proteins 1 and 2 (TYRP1, TYRP2). In contrast to alpha-MSH, BA activated protein kinase A in a cAMP independent manner. Also, para-topolin induced tyrosinase expression and melanogenesis in B16 cells (our unpublished data).

14.4.4 Clinical Examination of Kinetin and Its Relatives

Topical K has been evaluated in multiple open-label single-arm clinical trials, in which effects were compared before and after treatment. K lotion 0.1% (Kinerase) applied twice daily was evaluated as a treatment for mildly to moderately photodamaged facial skin in 32 subjects. Twelve and twenty four week treatments were found to improve significantly skin texture, mottled hyperpigmentation and fine wrinkles according to the both patient’s and physician’s assessment. The treatment also improved transdermal water loss (McCullough and Weinstein 2002). The same regimen was later evaluated in patients (N = 17) with mild to moderate facial rosacea in a 12 week study (Wu et al. 2007). K treatment was found to reduce redness and had also a positive, yet statistically insignificant, effect on telangiec- tasia. The physician’s rating of overall improvement of rosacea symptoms increased over the study period and by week 12, almost 60% of the subjects showed at least moderate improvement. K was generally well tolerated in those studies. However, in rare cases, acne or a rash was observed after 8 weeks’ treatment. Wanitphakdeedecha et al. (2015) evaluated the effects of applying 0.1% K cream 320 J. Voller et al. twice daily on the photoaging facial skin of 100 Thai subjects. Twelve weeks’ treatment was accompanied by small but statistically significant improvements in overall skin condition, skin texture, color and wrinkles. K also improved ultraviolet spots and redness. Chiu et al. (2007) evaluated possible synergistic effects of using a combination of K (0.03%) and niacin (4%) in a double-blind clinical study on 52 Taiwanese subjects. Serum containing either K and niacin (27 subjects) or niacin alone (25 subjects) was applied to one half of the face and the vehicle to the other twice daily for 12 weeks. Although the combination of K and niacin had a larger positive effect on most of the evaluated parameters, including corneal hydration and erythema index, than niacin alone when compared to the baseline, the differences between the treatments were not statistically significant. Also, vehicle alone had a positive effect on multiple parameters, yet the effects of K were typically larger. To the best of our knowledge, no reports of clinical trials with tZ have been published. Several open-label, single-arm clinical trials have evaluated the K derivative 6-furfurylamino-9-(tetrahydropyran-2-yl)-9H-purine (Pyratine-6, PRK-124), which originated from our laboratory (Laboratory of Growth Regulators, Olomouc, Czech Republic) and then was developed by companies Senetek PLC and Pyratine PLC, USA. In a trial evaluating its effects on aging and photodamaged skin of 40 subjects (34 finished the study), Pyratine (0.1%) was reported to improve skin moisturization, roughness, mottled hyperpigmentation, fine wrinkles and facial erythema in comparison with the baseline within 4 weeks (McCullough et al. 2008). Application of the compound (0.125%) twice daily was also found to improve symptoms of acne rosacea compared to the baseline in a 12 week study (Ortiz et al. 2009), with some beneficial effects observed as early as week 4. In a later study, 18 subjects were followed over an extended 48 week trial (Tremaine et al. 2010). A mean 44% reduction in erythema severity and 89% reduction in inflammatory lesions was observed at week 48. The treatment also had a positive effect on telangiectasias, transepidermal water loss and skin dryness. No treatment-induced skin irritation was observed.

14.5 Neuroprotective Activity of Cytokinins

Multiple reports of cytokinin activities relevant to the treatment of neurodegeneration exist. Notably, such effects are not limited to cytokinin bases only, but cytokinin ribosides have also been shown to possess neuroprotective activity. As discussed above, K protects rat astrocytes in vitro as well as rat brain against glycoxidative damage in the D-galactose model via promotion of anti-oxidant defense (Liu et al. 2011). KRTP increases kinase activity of PTEN-induced putative kinase 1 (PINK1). PINK1 is critical for mitochondrial quality control—it accumulates in the outer membrane of impaired mitochondria and, through the recruitment of Parkin protein, targets them for autophagy. Both PINK1 and Parkin are mutated in the familial forms of recessive Parkinson’s disease. Treatment with K protected SH-SY5Y5 neuroblastoma cells against proteasomal and oxidative 14 Plant Hormone Cytokinins for Modulating Human Aging … 321 stress in PINK1-dependent manner. K is therefore an interesting drug candidate for the treatment of familial form of Parkinson’s disease and possibly also therapy of other diseases with mitochondrial dysfunction. It would be interesting to identify other N6-substitued adenines with similar type of activity. This may prove difficult because both base and riboside require not only multistep metabolic activation, but they also have to mimic the shape and behavior of ATP in the PINK1 binding pocket (Hertz et al. 2013). K is also a candidate drug for neurodegenerative disease familial dysautonomia. A clinical study (NCT02274051) is in the process of recruiting patients at the time of writing this text according to web page www.clinicaltrials.gov. Both in vitro and in vivo experiments have demonstrated that K is able to correct aberrant splicing of pre-mRNA originating from the IKBKAP gene (inhibitor of kappa light polypep- tide gene enhancer in B-cells, kinase complex-associated protein). The mechanism of K’s action is unknown. However, the very limited number of transcripts with splicing influenced by the treatment suggests that K interacts with regulators or components of specific spliceosome sub-species. tZ also exhibits multiple activities relevant for the therapy of diseases of the central nervous system, in particular Alzheimer’s disease. It was identified as the substance responsible for inhibition of rat acetylcholinesterase contained in an −4 extract from the traditional Korean edible plant Fiatoua villosa (IC50 1.09 Â 10 M) (Heo et al. 2002). Follow-up studies showed that tZ protects rat pheocytohroma cells PC12 against toxic effects of an amyloid beta fragment comprising amino acids 25–35. tZ treatment had a positive effect on both ROS production and cell viability (Choi et al. 2009). Also reported was a beneficial effect of long-term treatment with tZ on scopolamine-induced amnesia. Scopolamine is an antagonist of muscarinic acetylcholine receptors, but its administration also influences other neurotransmitter systems. For example, repeated treatment of scopolamine has been shown to decrease levels of monoamines (noradrenaline, dopamine and serotonin) and also increase oxidative stress in brain (Haider et al. 2016). In Choi et al.’s study (2009), animals had ad libitum access to either normal drinking water (control and scopolamine control groups) or tZ solution (0.002, 0.004, and 0.008% w/v) for 21 days. More specific information about dosing (drinking volume by the individual animals) and tZ concentrations in plasma and brain tissues was not reported. After the treatment period, temporal amnesia was induced by a single s.c. injection of scopolamine (1 mg/kg) and behavioral tests were started 30 min later. All three tested concentrations markedly attenuated negative effects of scopolamine on passive avoidance and spontaneous alternation behaviors in the Y-maze test. After finishing the tests, the animals were sacrificed and acetylcholinesterase activity in brain lysates was measured. The two highest tZ concentrations were able to reduce the effects of scopolamine on acetylcholinesterase activity. Similar effects on behavior were observed when mice were provided food mixed with tZ ad libitum for 4 weeks (Choi et al. 2009). However, the mechanism of the observed tZ activity is unclear. The authors provided no rationale for the treatment regime length— much shorter regimes and possibly single tZ injection may have been more appropriate for evaluation of acetylcholinesterase inhibition or direct antioxidant 322 J. Voller et al. capacity. It would be interesting to assess the effect of tZ on the level of oxidative stress and level/activity of the antioxidant defense system in brain before and after scopolamine treatment. KR and trans-zeatin ribosides (tZRs) have been shown to have cytoprotective activities. Both of them, but not BAR, can protect PC12 cells against serum star- vation induced apoptosis (Lee et al. 2012). The corresponding free bases, tZ and K were inactive in this model. The protective effects of KR were apparent at concentrations as low as 1 lM, whereas tZR did not show any significant activity at this concentration. However, at a concentration of 100 lM, tZR was more effective than equimolar KR and was therefore selected for follow-up experiments. Its protective activity in a serum deprivation model was mediated by activation of the A2A adenosine receptor (A2AR) and the downstream protein kinase A (PKA) as co-treatment with inhibitors of those proteins abolished the protective effect. A2AR has been suggested as a potential therapeutic target in Huntington’s disease because it is highly expressed in the striatum, where mutant huntingtin causes early damage. Moreover, A2AR-selective agonists effectively ameliorate several symptoms of Huntington’s disease in both cell cultures and animal models (Chou et al. 2005; Chiu et al. 2015). Therefore, tZR’s ability to prevent huntingtin aggregation was tested in PC12 cells overexpressing the mutant protein. Indeed, tZR treatment prevented both the mutant huntingtin aggregation and subsequent proteasome dysfunction in an A2AR and PKA dependent manner. Para-Topolin riboside (pTR, 6-(4-hydroxybenzylamino)-9H-purine riboside) is an A2AR agonist as well. It was isolated as a neuroprotective substance from the Chinese medicinal plant Gastrodia elata (Huang et al. 2011a, b). pTR (designated as T1-11 in the study) has been shown to interact with not only A2AR but also equilibrative nucleotide transporter ENT1. ENT1 inhibition can increase the amount of adenosine in extracellular space and possibly contribute to modulation of adenosinergic signaling in synapses where both proteins are present. Treatment of R6/2 mice with pTR administered using a subcutaneous Alzet minipump for 48 h improved motor deterioration. pTR administered in drinking water (0.05 mg/ml ad libitum, plasmatic or brain concentration unknown) improved the rotarod performance already after 2 weeks and the effect persisted until the end of the experiment after 7 weeks. Analysis of the brains showed a lower accumulation of huntingtin, improved activity of proteasome and higher levels of mRNA for brain-derived neurotrophic factor (BDNF). Finally, pTR also binds to A3R, another adenosine receptor with a known role in CNS physiology (Borea et al. 2014). For example, agonists of A3R have been shown to have neuroprotective effects in subarachnoid hemorrhage-induced brain damage (Maria Pugliese et al. 2007). Nevertheless, the therapeutic utility of targeting the brain is somewhat controversial because prolonged A3R activation may be neurotoxic (Luo et al. 2010). Recently, A3R agonistic activity of iPR has been reported (Blad et al. 2011). However, to the best of our knowledge, its effects in models of neurological diseases have not yet been studied. 14 Plant Hormone Cytokinins for Modulating Human Aging … 323

14.6 Immunomodulatory Activities of Cytokinins

In the previous section, we discussed interaction of cytokinin ribosides with the adenosine receptor A2A in relation to their neuroprotective activities. However, A2A is also expressed in various immune system cells (T-lymphocytes, macrophages, monocytes and polymorphonuclear leukocytes) and its activation plays an important role in the regulation of both innate and adaptive immunity (Milne and Palmer 2011). Selective A2A receptor agonists have been suggested as drugs for treatment of graft-versus-host disease, colitis and T-lymphocyte mediated ischemia- reperfusion injury (Chhabra et al. 2012; Jones and Kang 2015). The immunomodulatory activity of tZR was demonstrated for the first time by Lappas (2015). It was shown to inhibit production of interferon (IFN)-c, IL-2 and TNF-a in either CD3+CD4+ or CD3+CD8+ T-lymphocytes incubated with anti-CD3 antibody. In the case of the CD3+CD4+ population, tZR also inhibited production of IL-4. EC50 values were in the range 10–100 lM. Furthermore, tZR inhibited upregulation of CD25, CD69 and CD40L. The inhibitory effects in the latter assays were strongly attenuated by co-treatment with 10 lM of the selective A2A receptor antagonist ZM241385, thereby confirming its involvement. In mice, tZR decreased the number of white-blood cells in intraperitoneal infiltrate in thioglycollate- induced peritonitis. tZR was administered as a 1 mg/kg i.p. bolus at times 0, 1.5 and 3 h after the thioglycollate injection. Overall, the data showed that tZR is not a very potent immunomodulator in terms of the active concentration. However, the absence of cytotoxic effects up to 1000 lM suggest that the low activity could be possibly compensated by the administration of higher doses. As discussed above, iPR and BAR activate the Nrf2 pathway and protect cells against oxidative stress (Dassano et al. 2014). The employed models also included 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced superoxide production by HL-60 cells differentiated along the neutrophilic lineage. Further, the authors tested the activity of the cytokinins in a mouse ear inflammation model where topical TPA-induced oxidative stress stimulates an inflammatory response. Pretreatment with iPR and BAR attenuated the inflammation and reduced the number of infil- trating neutrophils. The exact mechanism of the protective effect is unclear and other processes besides Nrf2 activation could be involved. The authors discussed possible involvement of glucocorticoid receptor signaling (iPR induced several transcripts in this pathway) and adenosine receptor A3R activation. The A3R agonistic activity of iPR and tZR was reported recently (Blad et al. 2011). A3R regulates various aspects of inflammation, including neutrophile chemotaxis and superoxide production (van der Hoeven et al. 2010). The anti-inflammatory effects of A3R activation are mediated through the inhibition of NF-jB dependent production of cytokines, including TNF-a. A3R agonists have been shown to have robust anti-inflammatory activity in animal models of inflammatory bowel disease, systemic toxemia, pulmonary and liver inflammation as well as rheumatoid arthritis (Borea et al. 2014). Nanomolar A3R agonist CF101 (IB-MECA) structurally related to aromatic cytokinins is currently being evaluated as an antirheumatic and 324 J. Voller et al. antipsoriatic agent in clinical trials. Recently, promising activity in patients with moderate to severe plaque psoriasis was reported (David et al. 2016). Another explanation of the anti-inflammatory activity of iPR was offered by studies of its effects on natural killer cells (NK cells) (Ciaglia et al. 2014). iPR at 10 lM concentration inhibited the cytotoxicity of NK cells against leukemia K562 cells. The treatment prevented ERK/MAPK and STAT5 activation in IL-2-activated NK cells, downregulated the expression of activating receptors NKp44 and NKG2D and decreased secretion of cyto/chemokines (RANTES, MIP-1a, TNF-a and IFN-c). Topical application of iPR significantly reduced ear edema in a mouse model of croton oil-induced ear dermatitis with a potency comparable to that of indomethacin. Histology analysis showed lower leukocyte infiltration and decreased staining of the natural cytotoxicity receptor NKp46, whose expression is typical for NK cells, in irritated mid and papillary dermis. Finally, cytokinin bases may also possess anti-inflammatory activity. K and Pyratine have been shown to have beneficial effects on acne rosacea, a skin condition with an inflammatory component (Wu et al. 2007; Tremaine et al. 2010).

14.7 Antiproliferative Effects of Cytokinins and Cytokinin Analogues

Natural cytokinin ribosides iPR, KR, BAR, ortho-topolin riboside (oTR) and N6- (2-hydroxy-3-methoxybenzyl)adenosine (but not their respective bases) have been reported to exhibit strong cytotoxic effects against a range of human cell lines derived from both hematological malignancies and solid tumors (Doležal et al. 2007; Voller et al. 2010, 2017). Numerous studies with various experimental designs (assay principle, endpoint, length of treatment) have demonstrated that some cytokinin ribosides are active at submicromolar (against some leukemia) or micromolar concentrations (against other leukemia, adherent cells). The toxicity of tZR and cis-zeatin riboside, other natural cytokinins that differ from iPR by hydroxylation of the isoprenoid side-chain, is very limited (Ishii et al. 2002; Rattan and Sodagam 2005; Voller et al. 2010). Further, isomers of oTR with a hydroxy group in either the meta-orpara-position of the phenyl ring do not show signiﬁcant toxicity (Voller et al. 2010). In leukemia cell lines, cytokinin ribosides induce rapid apoptosis (Mlejnek and Kuglík 2000; Ishii et al. 2002; Voller et al. 2017). Cell death is preceded by ATP depletion. An exception is N6-(2-hydroxy-3-methoxybenzyl)adenosine, where apoptosis ensues without marked effects on cell energy levels (Voller et al. 2017). It has recently been demonstrated that KR may be a potential drug for the treatment of multiple myelomas. In several cell lines, KR has been found to suppress cyclin D1 and D2 transcription, followed by arrest of the cell-cycle and selective apoptosis in tumor cells (Tiedemann et al. 2008). KR may be also effective against leukemia stem cells (McDermott et al. 2012). 14 Plant Hormone Cytokinins for Modulating Human Aging … 325

Cytotoxic effects of iPR and KR against mammalian cell lines derived from solid tumors have been reported many times (Meisel et al. 1998; Griffaut et al. 2004; Laezza et al. 2006, 2009, 2010; Spinola et al. 2007; Cheong et al. 2009; Cabello et al. 2009; Colombo et al. 2009; Rajabi et al. 2012; Wang et al. 2012; Ciaglia et al. 2014, 2016). Depending on the cell line and cytokinin used, the treatment resulted in apoptosis, G1 or G2/M block. The spectrum of the effects induced by cytokinin ribosides in the tested cell lines included ATP depletion, genotoxic stress (Cabello et al. 2009), JNK activation (Laezza et al. 2009), inhibition of farnesyl-protein transferase activity (Laezza et al. 2006), inhibition of EGFR signaling (Ciaglia et al. 2016) and changes in levels of mitochondrial proteins (Cheong et al. 2009). Recently, microarray analysis of the effects of iPR (100 lM) on MCF7 and A549 cell lines was published. iPR induced a set of genes involved in stress induced cell cycle arrest, e.g., PPP1R15A, DNAJB9, DDIT3, and HBP1 (Colombo et al. 2009). Cytokinin ribosides may also interfere with neoangiogenesis (Pisanti et al. 2014). Further, cytokinin riboside-5′-monophosphates have been identified as inhibitors of putative oncogene RCL1 (Amiable et al. 2013; Voller et al. 2017). In vivo anticancer activity of iPR, KR, oTR and BAR has been demonstrated using several animal and xenograft models of cancer (Griffaut et al. 2004; Laezza et al. 2006; Tiedemann et al. 2008; Voller et al. 2010). iPR and BAR have been shown to exhibit limited activity against a diverse range of cancers in a small clinical trial (Mittelman et al. 1975). Micromolar concentrations of both cytokinin ribosides and cytokinin bases can also induce cell death in plant cell cultures, with some traits typical for apoptosis (activation of caspase-like proteases and fragmentation of DNA) (Mlejnek and Procházka 2002; Mlejnek and Doležel 2005). This cell death is preceded by depletion of ATP and the production of ROS. In contrast to their hormonal activity in plants, which requires interaction with specific membrane-bound receptors, intracellular conversion of cytokinins to 5′- monophosphates is necessary for their cytotoxic effect. The concentrations of cytokinins required to produce cytotoxic effects are higher than those found endogenously in plant tissues, but they do fall within the range used in plant bioassays (Carimi et al. 2003; Mlejnek and Doležel 2005). Phosphorylation of cytokinin ribosides by adenosine kinase (ADK) is a requirement for the cytotoxic effect of cytokinin ribosides in both animal (Mlejnek and Doležel 2005; Voller et al. 2010) and plant cells (Mlejnek and Procházka 2002). Low affinity to ADK (and possibly other nucleoside kinases) may explain the lack of activity of other cytokinin ribosides (Mlejnek and Doležel 2005) and possibly also their analogues with ribose replaced by acyclic polyols (Colombo et al. 2009; Ottria et al. 2009). Contrary to other nucleoside analogues that are converted to nucleoside triphos- phates, the dominant metabolites of cytokinin ribosides are their respective riboside monophosphates (Mlejnek and Doležel 2005). This observation suggests that cytokinin ribosides have a different mechanism of action than classical antimetabolites which, after phosphorylation, directly interfere with the synthesis of nucleic acids. In cultured plant cells, the cytoxicity of cytokinin bases and their corresponding ribosides are comparable because, in contrast to human cell lines, 326 J. Voller et al. plant cells are able to convert both metabolic forms efficiently into riboside-5 -monophosphateś (Mlejnek and Procházka 2002; Mlejnek and Doležel. 2005). A characteristic trait of leukemia cells is blockade of their differentiation into functional mature cells. Various chemicals acting by diverse mechanisms are able to force malignant cells to undergo terminal differentiation. Such differentiation therapy could be much safer than regimens based on cytotoxic effects. Cytokinin bases K, iP, BA and, to lesser degree, also tZ have been shown to induce granulocytic differentiation of human myeloid leukemia cell line HL-60 (Ishii et al. 2002) derived from the peripheral blood leukocytes of a patient with acute myeloid leukemia. The activity was mediated by phosphorylation of ERK1/2, expression of CEBPD and S100P (Ishii et al. 2005a, b). Whereas cytokinin bases induced differentiation at rather high concentrations (25–100 lM), their ribosides caused rapid apoptosis at low micromolar levels. Treatment with caspase inhibitors shifted the activity of iPR in HL-60 from pro-apoptotic to growth inhibitory and differentiating activity (Ishii et al. 2002). Cytokinins are also able to promote differentiation of keratinocytes, as discussed above.

14.8 Conclusions

Cytokinins regulate a wide range of plant processes, including growth, differentiation and organ senescence. This fascinating activity together with reports that some cytokinins occur in animals, including humans, has inspired interest in studying their effects in mammalian cell cultures and animal models. However, current knowledge of cytokinin signaling in plants suggests that the diverse effects of cytokinins in animal systems are relayed by mechanisms and pathways different from those that mediate hormonal activity in plants. In plants, cytokinins are recognized by receptor histidine kinases, which are part of the His-Asp phosphorelay resembling the two-component environmental sensors of bacteria. No signaling system with similar organization exists in animals. In addition, plant leaf senescence, an active and highly regulated process, is a phe- nomenon completely different from animal aging (stochastic accumulation of damage) and cell senescence (irreversible growth arrest that protects the body against tumor development). Nevertheless, studies of cytokinins in mammalian cell cultures and animals have led to discoveries of a range of responsive pathways and processes, as well as numerous prospective therapeutic applications. The propensity of cytokinins, which are adenine and adenosine derivatives, to influence diverse processes in animal cells is probably a consequence of their ability to interact with various components of the animal purinome. The anti-proliferative activity of cytokinin ribosides (through induction of cell cycle block or/and cell death) and bases (through induction of cell differentiation) has prompted studies into their potential utility for the therapy of proliferative diseases like leukemias, cancers and psoriasis. Although anti-cancer cytokinin 14 Plant Hormone Cytokinins for Modulating Human Aging … 327 ribosides were evaluated in patients with diverse malignancies as early as the 1970s (Mittelman et al. 1975), limited activity and problems with stability led to a loss of interest in further development at that time. Recent interest in cytotoxic anti-cancer ribosides has been stimulated by molecular studies of their mechanism of action, e.g., by microarray analyses (Colombo et al. 2009) and discovery of the high anticancer activity of BARs hydroxylated on the phenyl ring (Doležal et al. 2007; Voller et al. 2010). Further, inhibitors of cyclin-dependent kinases olomoucine, bohemine, roscovitine, developed in our laboratory, were inspired by cytokinins (for a review see Jorda et al. 2012). Another fertile research area was spurred on by the finding that the cytokinin bases K and tZ delay the onset of several characteristics related to aging in human skin fibroblasts during serial passage in vitro (Rattan and Clark 1994; Rattan and Sodagam 2005). Some cytokinin bases even extend the lifespan of invertebrates such as drosophila (Sharma et al. 1995) and C. elegans (our unpublished data). Even today, when tZ, K and its analogue Pyratine are the principal components of cosmetics used for the treatment of aging skin, their mechanism of action is still not fully understood. Since the original discovery, research has mainly focused on the ability of cytokinins to protect macromolecules and cells against oxidative damage and stress. Numerous studies have reported the ability of K to quench radicals directly and this activity has also been observed for several other cytokinin bases. However, K and tZ also induce cellular antioxidant defense by an as yet uniden- tified mechanism. Moreover, K modulates splicing of certain pre-RNAs (Slaugenhaupt et al. 2004). Therefore, it is tempting to speculate that it may also modulate splicing of transcripts of some genes related to cytoprotection. Recently, Hertz et al. (2013) described an unusual mechanism of cytoprotective activity of K, including its intracellular conversion into respective ribotides. KRTP increases the activity of the mitoprotective kinase PINK1 by acting as a more active ATP analogue (neosubstrate). Other mechanisms of action including the ribosyla- tion of cytokinin bases cannot be ruled out either. Cytokinin ribosides induce Nrf2-dependent transcription (Dassano et al. 2014, our unpublished data), inhibit proteosynthesis (as indicated by a decrease of activating phosphorylation of p70 S6 kinase and S6 protein) and activate autophagy (increase in levels of LC3-II) (our unpublished data). All these pathways have been implicated in cytoprotection and aging prevention (Bruns et al. 2015; Madeo et al. 2015). If cytokinin bases are metabolized into respective ribosides/ribotides in quantities sufficient to activate those stress response pathways, they could act as hormetin precursors, i.e., prohormetins. In the last few years, cytokinin ribosides have been studied in connection with their anti-inflammatory and neuroprotective activities. Activation of A2A or A3 adenosine receptors has been implicated in the effect. Because much more active agonists have already reached clinical trials, purinergic activity alone may not warrant future development of such compounds into drugs. However, other pathways could possibly be involved in immunomodulatory activity, including 328 J. Voller et al. activation of an Nrf2 response and inhibition of farnesyl diphosphate synthase. Notably, inhibition of protein farnesylation by iPR mediates not only its anti-cancer and immunomodulatory activities (Ciaglia et al. 2014; Laezza et al. 2006) but also its ability to decrease the accumulation of abnormal lamin A (progerin) in the nuclear envelope of cells originating from patients with Hutchinson-Gilford progeria (Bifulco et al. 2013). Progerin resulting from aberrant processing of wild type lamin A transcripts accumulates in tissues including skin of elderly and was therefore suggested as a biomarker of aging (McClintock et al. 2007). Hence, topical application of iPR (or possibly iP) could have multiple benefits. A major obstacle in the development of cytokinin ribosides into anticancer drugs has been their pharmacokinetics. Clinical trials in the 1970s showed that they have a short plasmatic half-life (Mittelman et al. 1975). Our studies indicate that cytokinin ribosides may have problems with crossing biological barriers (Voller et al. 2010). However, studies in rodents reviewed above suggest that some cytokinin ribosides can be absorbed from the gut and even reach the brain, whereas others are able to penetrate the skin, as indicated by their therapeutic effects. Unfortunatelly, no information about plasma and/or tissue concentration has been reported. Such observations do not guarantee similar behavior in humans because of differences in physiology. For example, human skin has not only less hair follicles but also a thicker epidermis and dermis, which pose a much higher barrier to the penetration of drugs, especially hydrophilic ones. Prodrugs may be necessary to allow the active compound to reach target deep layers of (intact) skin. Another option could be administration of cytokinin bases that could be converted into active ribosides within the tissue, as discussed above. Once a cytokinin riboside is present in target skin layers, its pharmacokinetic properties may be seen as an advantage as they will prevent it from reaching the systemic circulation, and thus limit possible side effects (e.g., vasodilatation in the case of A2AR agonists). K, tZ and Pyratine are principle components of cosmetics products (Kinerase and Pyratine-6 lines). Clinical evaluation reports have been published for K and Pyratine. Both compounds showed beneficial effects on photoaging skin and also ameliorated some symptoms of acne rosacea. Most of the studies were open-label and single-arm and effects were compared to the state before the treatment. The reported positive results together with multiyear experience of their safe usage in cosmetics could encourage testing of these compounds in clinical trials with a better design that may eventually support their approval in acne rosacea or possibly other inflammatory skin conditions. Recent discoveries of the unique activities of K relevant for the treatment of Parkinson’s disease (PINK1 activation) or familial dysautonomia (correction of aberrant splicing of IKBKAP transcripts) may result in clinical trials in those indications. Recruitment of patients for a study of familial dysautonomia is currently underway. Because the therapy would require chronic administration of K, we may also learn about its anti-aging activities not yet described from those studies. 14 Plant Hormone Cytokinins for Modulating Human Aging … 329

Acknowledgements The study was partly supported by the Internal Grant Agency of Palacký University (IGA_PrF_2017_013), Endowment Fund of Palacký University and by the Grant Agency of the Czech Republic (7-14007S). This work was also funded by the Ministry of Education, Youth and Sports of the Czech Republic (the National Program for Sustainability I: Nr. LO1204). The authors also would like to thank Sees-editing Ltd. (UK) for the English editing of this manuscript.

References

Amiable C, Pochet S, Padilla A, Labesse G, Kaminski PA (2013) N6-substituted AMPs inhibit mammalian deoxynucleotide N-hydrolase DNPH1. PLoS ONE 8:e80755. doi:10.1371/journal. pone.0080755 Barciszewski J, Siboska GE, Pedersen BO, Clark BFC, Rattan SIS (1996) Evidence for the presence of kinetin in DNA and cell extracts. FEBS 393:197–200 Barciszewski J, Siboska GE, Pedersen BO, Clark BFC, Rattan SIS (1997) A mechanism for the in vivo formation of N6-furfuryladenine, kinetin, as a secondary oxidative damage product of DNA. FEBS Lett 414:457–460. doi:10.1016/S0014-5793(97)01037-5 Barciszewski J, Mielcarek M, Stobiecki M, Siboska G, Clark BFC (2000) Identification of 6-furfuryladenine (kinetin) in human urine. Biochem Biophys Res Commun 279:69–73. doi:10.1006/bbrc.2000.3928 Berge U, Kristensen P, Rattan SIS (2006) Kinetin-induced differentiation of normal human keratinocytes undergoing aging in vitro. Ann NY Acad Sci 1067:332–336. doi:10.1196/annals. 1354.045 Berge U, Kristensen P, Rattan SIS (2008) Hormetic modulation of differentiation of normal human epidermal keratinocytes undergoing replicative senescence in vitro. Exp Gerontol 43:658–662. doi:10.1016/j.exger.2007.12.009 Bifulco M, D’Alessandro A, Paladino S, Malfitano AM, Notarnicola M, Caruso MG, Laezza C (2013) N6-isopentenyladenosine improves nuclear shape in fibroblasts from humans with progeroid syndromes by inhibiting the farnesylation of prelamin A. FEBS J 280:6223–6232. doi:10.1111/febs.12544 Blad CC, von Frijtag Drabbe Künzel JK, de Vries H, Mulder-Krieger T, Bar-Yehuda S, Fishman P, Ijzerman AP (2011) Putative role of the adenosine A(3) receptor in the antiproliferative action of N (6)-(2-isopentenyl)adenosine. Purinergic Signal 7:453–462. doi:10.1007/s11302-011-9244-9 Borea PA, Varani K, Vincenzi F, Baraldi PG, Tabrizi MA, Merighi S, Gessi S (2014) The A3 adenosine receptor: history and perspectives. Pharmacol Rev 67:74–102. doi:10.1124/pr.113. 008540 Brizzolari A, Marinello C, Carini M, Santaniello E, Biondi PA (2016) Evaluation of the antioxidant activity and capacity of some natural N6-substituted adenine derivatives (cytokinins) by fluorimetric and spectrophotometric assays. J Chromatogr B: Anal Technol Biomed Life Sci 1019:164–168. doi:10.1016/j.jchromb.2015.12.047 Bruns DR, Drake JC, Biela LM, Peelor FF, Miller BF, Hamilton KL (2015) Nrf2 signaling and the slowed aging phenotype: evidence from long-lived models. Oxid Med Cell Longev 2015:1–15. doi:10.1155/2015/732596 Cabello CM, Bair WB 3rd, Ley S, Lamore SD, Azimian S, Wondrak GT (2009) The experimental chemotherapeutic N6-furfuryladenosine (kinetin-riboside) induces rapid ATP depletion, genotoxic stress, and CDKN1A(p21) upregulation in human cancer cell lines. Biochem Pharmacol 77:1125–1138. doi:10.1016/j.bcp.2008.12.002 330 J. Voller et al.

Carimi F, Zottini M, Formentin E, Terzi M, Lo Schiavo F (2003) Cytokinins: new apoptotic inducers in plants. Planta 216:413–421. doi:10.1007/s00425-002-0862-x Chen C, Kristopeit SM (1981) Metabolism of cytokinin: deribosylation of cytokinin ribonucle- oside by adenosine nucleosidase from wheat germ cells. Plant Physiol 68:1020–1023. doi:10.1104/pp.68.5.1020 Cheong J, Goh D, Yong JWH, Tan SN, Ong ES (2009) Inhibitory effect of kinetin riboside in human heptamoa, HepG2. Mol BioSyst 5:91–98. doi:10.1039/b712807j Chhabra P, Linden J, Lobo P, Okusa MD, Brayman KL (2012) The immunosuppressive role of adenosine A2A receptors in ischemia reperfusion injury and islet transplantation. Curr Diabetes Rev 8:419–433 Chheda G, Mittelman A (1972) N6-(D2-isopentenyl) adenosine metabolism in man. Biochem Pharmacol 21:27–37 Chiu P-C, Chan C-C, Lin H-M, Chiu H-C (2007) The clinical anti-aging effects of topical kinetin and niacinamide in Asians: a randomized, double-blind, placebo-controlled, split-face comparative trial. J Cosmet Dermatol 6:243–249. doi:10.1111/j.1473-2165.2007.00342.x Chiu F-L, Lin J-T, Chuang C-Y, Chien T, Chen C-M, Chen K-H, Hsiao H-Y, Lin Y-S, Chern Y, Kuo H-C (2015) Elucidating the role of the A2A adenosine receptor in neurodegeneration using neurons derived from Huntington’s disease iPSCs. Hum Mol Genet 24:6066–6079. doi:10. 1093/hmg/ddv318 Choi SJ, Jeong C-H, Choi S-G, Chun J-Y, Kim YJ, Lee J, Shin D-H, Heo HJ (2009) Zeatin prevents amyloid beta-induced neurotoxicity and scopolamine-induced cognitive deficits. J Med Food 12:271–277. doi:10.1089/jmf.2007.0678 Chou S-Y, Lee Y-C, Chen H-M, Chiang M-C, Lai H-L, Chang H-H, Wu Y-C, Sun C-N, Chien C-L, Lin Y-S, Wang S-C, Tung Y-Y, Chang C, Chern Y (2005) CGS21680 attenuates symptoms of Huntington’s disease in a transgenic mouse model. J Neurochem 93:310–320. doi:10.1111/j.1471-4159.2005.03029.x Ciaglia E, Pisanti S, Picardi P, Laezza C, Sosa S, Tubaro A, Vitale M, Gazzerro P, Malfitano AM, Bifulco M (2014) N6-isopentenyladenosine affects cytotoxic activity and cytokines production by IL-2 activated NK cells and exerts topical anti-inflammatory activity in mice. Pharmacol Res 89:1–10. doi:10.1016/j.phrs.2014.07.003 Ciaglia E, Abate M, Laezza C, Pisanti S, Vitale M, Seneca V, Torelli G, Franceschelli S, Catapano G, Gazzerro P, Bifulco M (2016) Antiglioma effects of N6-isopentenyladenosine, an endogenous isoprenoid end product, through the downregulation of epidermal growth factor receptor. Int J Cancer. doi:10.1002/ijc.30505 Colombo F, Falvella FS, De Cecco L, Tortoreto M, Pratesi G, Ciuffreda P, Ottria R, Santaniello E, Cicatiello L, Weisz A, Dragani TA (2009) Pharmacogenomics and analogues of the antitumour agent N6-isopentenyladenosine. Int J Cancer 124:2179–2185. doi:10.1002/ijc.24168 Dassano A, Mancuso M, Giardullo P, De Cecco L, Ciuffreda P, Santaniello E, Saran A, Dragani TA, Colombo F (2014) N(6)-isopentenyladenosine and analogs activate the NRF2-mediated antioxidant response. Redox Biol 2:580–589. doi:10.1016/j.redox.2014.03.001 David M, Gospodinov DK, Gheorghe N, Mateev GS, Rusinova MV, Hristakieva E, Solovastru LG, Patel RV, Giurcaneanu C, Hitova MC, Purcaru AI, Horia B, Tsingov II, Yankova RK, Kadurina MI, Ramon M, Rotaru M, Simionescu O, Benea V, Demerdjieva ZV, Cosgarea MR, Morariu HS, Michael Z, Cristodor P, Nica C, Silverman MH, Bristol DR, Harpaz Z, Farbstein M, Cohen S, Fishman P (2016) Treatment of plaque-type psoriasis with oral CF101: data from a phase II/III multicenter, randomized, controlled trial. J Drugs Dermatol 15:931–938 Doležal K, Popa I, Hauserová E, Spíchal L, Chakrabarty K, Novák O, Kryštof V, Voller J, Holub J, Strnad M (2007) Preparation, biological activity and endogenous occurrence of N6-benzyladenosines. Bioorg Med Chem 15:3737–3747. doi:10.1016/j.bmc.2007.03.038 Gan S, Amasino RM (1995) Inhibition of leaf senescence by autoregulated production of cytokinin. Science 270:1986–1988. doi:10.1126/science.270.5244.1986 14 Plant Hormone Cytokinins for Modulating Human Aging … 331

Ge L, Yong JWH, Tan SN, Yang XH, Ong ES (2004) Analysis of some cytokinins in coconut (Cocos nucifera L.) water by micellar electrokinetic capillary chromatography after solid-phase extraction. J Chromatogr A 1048:119–126 Gillissen B, Burkle L, Andre B, Kuhn C, Rentsch D, Brandl B, Frommer WB (2000) A new family of high-affinity transporters for adenine, cytosine, and purine derivatives in Arabidopsis. Plant Cell 12:291–300. doi:10.1105/tpc.12.2.291 Goldstein S, Czapski G (1991) SOD-like activity studies of cytokinin-copper(II) complexes. Free Radic Res Commun 12–13(Pt 1):173–177. doi:10.3109/10715769109145783 Griffaut B, Bos R, Maurizis J-CC, Madelmont J-CC, Ledoigt G (2004) Cytotoxic effects of kinetin riboside on mouse, human and plant tumour cells. Int J Biol Macromol 34:271–275. doi:10. 1016/j.ijbiomac.2004.06.004 Haider S, Tabassum S, Perveen T (2016) Scopolamine-induced greater alterations in neurochem- ical profile and increased oxidative stress demonstrated a better model of dementia: a comparative study. Brain Res Bull 127:234–247. doi:10.1016/j.brainresbull.2016.10.002 Havlíček L, Hanuš J, Veselý J, Leclerc S, Meijer L, Shaw G, Strnad M (1997) Cytokinin-derived cyclizn-dependent kinase inhibitors: synthesis and cdc2 inhibitory activity of olomoucine and related compounds. J Med Chem 40:408–412. doi:10.1021/jm960666x Heo H-J, Hong S-C, Cho H-Y, Hong B, Kim H-K, Kim E-K, Shin D-H (2002) Inhibitory effect of zeatin, isolated from Fiatoua villosa, on acetylcholinesterase activity from PC12 cells. Mol Cells 13:113–117 Hertz NT, Berthet A, Sos ML, Thorn KS, Burlingame AL, Nakamura K, Shokat KM (2013) A neo-substrate that amplifies catalytic activity of Parkinson’s-disease-related kinase PINK1. Cell 154:737–747. doi:10.1016/j.cell.2013.07.030 Hirose N, Takei K, Kuroha T, Kamada-Nobusada T, Hayashi H, Sakakibara H (2008) Regulation of cytokinin biosynthesis, compartmentalization and translocation. J Exp Bot 59:75–83. doi:10. 1093/jxb/erm157 Horgan R, Hewett EW, Horgan JM, Purse J, Wareing PF (1975) A new cytokinin from Populus x robusta. Phytochemistry 14:1005–1008. doi:10.1016/0031-9422(75)85176-4 Hsiao G, Shen M-Y, Lin K-H, Chou C-Y, Tzu N-H, Lin C-H, Chou D-S, Chen T-F, Sheu J-R (2003) Inhibitory activity of kinetin on free radical formation of activated platelets in vitro and on thrombus formation in vivo. Eur J Pharmacol 465:281–287. doi:10.1016/S0014-2999(03) 01528-0 Huang C-L, Yang J-M, Wang K-C, Lee Y-C, Lin Y-L, Yang Y-C, Huang N-K (2011a) Gastrodia elata prevents huntingtin aggregations through activation of the adenosine A2A receptor and ubiquitin proteasome system. J Ethnopharmacol 138:162–168. doi:10.1016/j.jep.2011.08.075 Huang N-K, Lin J-H, Lin J-T, Lin C-I, Liu EM, Lin C-J, Chen W-P, Shen Y-C, Chen H-M, Chen J-B, Lai H-L, Yang C-W, Chiang M-C, Wu Y-S, Chang C, Chen J-F, Fang J-M, Lin Y-L, Chern Y (2011b) A new drug design targeting the adenosinergic system for Huntington’s disease. PLoS ONE 6:e20934. doi:10.1371/journal.pone.0020934 Ishii Y, Hori Y, Sakai S, Honma Y (2002) Control of differentiation and apoptosis of human myeloid leukemia cells by cytokinins and cytokinin nucleosides, plant redifferentiation-inducing hormones. Cell Growth Differ 13:19–26 Ishii Y, Kasukabe T, Honma Y (2005a) Induction of CCAAT/enhancer binding protein-d by cytokinins, but not by retinoic acid, during granulocytic differentiation of human myeloid leukaemia cells. Br J Haematol 128:540–547. doi:10.1111/j.1365-2141.2004.05326.x Ishii Y, Kasukabe T, Honma Y (2005b) Immediate up-regulation of the calcium-binding protein S100P and its involvement in the cytokinin-induced differentiation of human myeloid leukemia cells. Biochim Biophys Acta Mol Cell Res 1745:156–165. doi:10.1016/j.bbamcr.2005.01.005 Ji C, Yang Y, Yang B, Xia J, Sun W, Su Z, Yu L, Shan S, He S, Cheng L, Wan Y, Bi Z (2010) Trans-Zeatin attenuates ultraviolet induced down-regulation of aquaporin-3 in cultured human skin keratinocytes. Int J Mol Med 26:257–263. doi:10.3892/ijmm-00000460 Jones KR, Kang EM (2015) Graft versus host disease: new insights into A2A receptor agonist therapy. Comput Struct Biotechnol J 13:101–105. doi:10.1016/j.csbj.2014.12.003 332 J. Voller et al.

Jorda R, Paruch K, Krystof V (2012) Cyclin-dependent kinase inhibitors inspired by roscovitine: purine bioisosteres. Curr Pharm Des 18:2974–2980 Kim S, Lee JJ, Jung E, Lee JJ, Huh S, Hwang H, Kim Y, Park D (2009) 6-Benzylaminopurine stimulates melanogenesis via cAMP-independent activation of protein kinase A. Arch Dermatol Res 301:253–258. doi:10.1007/s00403-008-0924-4 Kimura T, Doi K (2004) Depigmentation and rejuvenation effects of kinetin on the aged skin of hairless descendants of Mexican hairless dogs. Rejuvenation Res 7:32–39. doi:10.1089/ 154916804323105062 Kryštof V, Lenobel R, Havlíček L, Kuzma M, Strnad M (2002) Synthesis and biological activity of olomoucine II. Bioorg Med Chem Lett 12:3283–3286 Kudo T, Kiba T, Sakakibara H (2010) Metabolism and long-distance translocation of cytokinins. J Integr Plant Biol 52:53–60. doi:10.1111/j.1744-7909.2010.00898.x Kurakawa T, Ueda N, Maekawa M, Kobayashi K, Kojima M, Nagato Y, Sakakibara H, Kyozuka J (2007) Direct control of shoot meristem activity by a cytokinin-activating enzyme. Nature 445:652–655. doi:10.1038/nature05504 Laezza C, Notarnicola M, Caruso MG, Messa C, Macchia M, Bertini S, Minutolo F, Portella G, Fiorentino L, Stingo S, Bifulco M (2006) N6-isopentenyladenosine arrests tumor cell proliferation by inhibiting farnesyl diphosphate synthase and protein prenylation. FASEB J Off Publ Fed Am Soc Exp Biol 20:412–418. doi:10.1096/fj.05-4044lsf Laezza C, Caruso MG, Gentile T, Notarnicola M, Malfitano AM, Di Matola T, Messa C, Gazzerro P, Bifulco M (2009) N6-isopentenyladenosine inhibits cell proliferation and induces apoptosis in a human colon cancer cell line DLD1. Int J Cancer 124:1322–1329. doi:10.1002/ ijc.24056 Laezza C, Malfitano AM, Di Matola T, Ricchi P, Bifulco M (2010) Involvement of Akt/NF-jB pathway in N6-isopentenyladenosine-induced apoptosis in human breast cancer cells. Mol Carcinog 49:892–901. doi:10.1002/mc.20666 Lappas CM (2015) The plant hormone zeatin riboside inhibits T lymphocyte activity via adenosine A2A receptor activation. Cell Mol Immunol 12:107–112. doi:10.1038/cmi.2014.33 Lee Y-C, Yang Y-C, Huang C-L, Kuo T-Y, Lin J-H, Yang D-M, Huang N-K (2012) When cytokinin, a plant hormone, meets the adenosine A2A receptor: a novel neuroprotectant and lead for treating neurodegenerative disorders? PLoS ONE 7:e38865. doi:10.1371/journal.pone. 0038865 Liu Y, Zhang Z, Yang X (2011) Kinetin protects against lipid peroxidation and improves antioxidant status in cultured astrocytes and mouse brain exposed to D-galactose. Afric J Biotech 10:11721–11727. doi:10.5897/AJB11.289 Luo C, Yi B, Tao G, Li M, Chen Z, Tang W, Zhang JH, Feng H (2010) Adenosine A3 receptor agonist reduces early brain injury in subarachnoid haemorrhage. NeuroReport 21:892–896. doi:10.1097/WNR.0b013e32833dbd13 Madeo F, Zimmermann A, Maiuri MC, Kroemer G (2015) Essential role for autophagy in life span extension. J Clin Invest 125:85–93. doi:10.1172/JCI73946 Maria Pugliese A, Coppi E, Volpini R, Cristalli G, Corradetti R, Jeong LS, Jacobson KA, Pedata F (2007) Role of adenosine A3 receptors on CA1 hippocampal neurotransmission during oxygen–glucose deprivation episodes of different duration. Biochem Pharmacol 74:768–779. doi:10.1016/j.bcp.2007.06.003 McClintock D, Ratner D, Lokuge M, Owens DM, Gordon LB, Collins FS, Djabali K (2007) The mutant form of lamin A that causes Hutchinson-Gilford progeria is a biomarker of cellular aging in human skin. PLoS ONE 2:e1269. doi:10.1371/journal.pone.0001269 McCullough JL, Weinstein GW (2002) Clinical study of safety and efficacy of using topical kinetin 0.10% (Kinerase) to treat photodamaged skin. Cosmet Dermatol 15(9):29–32 McCullough JL, Garcia RL, Reece B (2008) A clinical study of topical Pyratine 6 for improving the appearance of photodamaged skin. J Drugs Dermatol JDD 7:131–135 McCullough JL, Weinstein GW (2002) Clinical study to assess the safety and efficacy of topical kinetin 0.10% (Kinerase®) for photodamaged skin. Cosmet Dermatol 15:29–32 14 Plant Hormone Cytokinins for Modulating Human Aging … 333

McDaniel DH, Neudecker BA, DiNardo JC, Lewis JA 2nd, Maibach HI (2005) Idebenone: a new antioxidant—Part I. Relative assessment of oxidative stress protection capacity compared to commonly known antioxidants. J Cosmet Dermatol 4:10–17. doi:10.1111/j.1473-2165.2005. 00152.x Mcdermott SP, Eppert K, Notta F, Isaac M, Datti A, Al-awar R, Minden MD, Dick JE, Dc W, Wrana J (2012) A small molecule screening strategy with validation on human leukemia stem cells uncovers the therapeutic efficacy of kinetin riboside A small molecule screening strategy with validation on human leukemia stem cells uncovers the therapeutic efficacy of ki. Blood 119:1200–1207. doi:10.1182/blood-2011-01-330019 Meisel H, Günther S, Martin D, Schlimme E (1998) Apoptosis induced by modified ribonucleosides in human cell culture systems. FEBS Lett 433:265–268 Milne GR, Palmer TM (2011) Anti-inflammatory and immunosuppressive effects of the A2A adenosine receptor. Sci World J 11:320–339. doi:10.1100/tsw.2011.22 Mittelman A, Evans JT, Chheda GB (1975) Cytokinins as chemotherapeutic agents. Ann NY Acad Sci 255:225–234 Mlejnek P, Doležel P (2005) Apoptosis induced by N6-substituted derivatives of adenosine is related to intracellular accumulation of corresponding mononucleotides in HL-60 cells. Toxicol In Vitro 19:985–990. doi:10.1016/j.tiv.2005.06.023 Mlejnek P, Kuglík P (2000) Induction of apoptosis in HL-60 cells by N6-benzyladenosine. J Cell Biochem 77:6–17. doi:10.1002/(SICI)1097-4644(20000401)77:1<6:AID-JCB2>3.0.CO;2-3 Mlejnek P, Procházka S (2002) Activation of caspase-like proteases and induction of apoptosis by isopentenyladenosine in tobacco BY-2 cells. Planta 215:158–166. doi:10.1007/s00425-002- 0733-5 Mok DWS, Mok MC (2001) Cytokinin metabolism and action. Annu Rev Plant Physiol Plant Mol Biol 52:89–118 Olsen A, Siboska GE, Clark BF, Rattan SI (1999) N(6)-Furfuryladenine, kinetin, protects against Fenton reaction-mediated oxidative damage to DNA. Biochem Biophys Res Commun 265:499–502. doi:10.1006/bbrc.1999.1669 Ortiz A, Elkeeb L, Truitt A, Hindiyeh R, Aquino L, Tran M, Weinstein G (2009) Topical PRK 124 (0.125%) lotion for improving the signs and symptoms of rosacea. J Drugs Dermatol JDD 8:459–462 Ottria R, Casati S, Maier JAM, Mariotti M, Ciuffreda P (2009) Novel isopentenyladenosine analogues: synthesis, characterization, and evaluation of antiproliferative activity on bladder carcinoma cells. nucleosides, nucleotides and nucleic acids 28:736–751. doi:10.1080/ 15257770903155550 Pisanti S, Picardi P, Ciaglia E, Margarucci L, Ronca R, Giacomini A, Malfitano AM, Casapullo A, Laezza C, Gazzerro P, Bifulco M (2014) Antiangiogenic effects of N6-isopentenyladenosine, an endogenous isoprenoid end product, mediated by AMPK activation. FASEB J 28:1132– 1144. doi:10.1096/fj.13-238238 Rajabi M, Gorincioi E, Santaniello E (2012) Antiproliferative activity of kinetin riboside on HCT-15 colon cancer cell line. Nucleosides Nucleotides Nucleic Acids 31:474–481. doi:10. 1080/15257770.2012.681825 Rattan SI, Clark BF (1994) Kinetin delays the onset of ageing characteristics in human fibroblasts. Biochem Biophys Res Commun 201:665–672 Rattan SIS, Sodagam L (2005) Gerontomodulatory and youth-preserving effects of zeatin on human skin fibroblasts undergoing aging in vitro. Rejuvenation Res 8:46–57. doi:10.1089/rej. 2005.8.46 Sharma SP, Kaur P, Rattan SI (1995) Plant growth hormone kinetin delays ageing, prolongs the lifespan and slows down development of the fruitfly Zaprionus paravittiger. Biochem Biophys Res Commun 216:1067–1071. doi:10.1006/bbrc.1995.2729 Sharma SP, Kaur J, Rattan SI (1997) Increased longevity of kinetin-fed Zaprionus fruitflies is accompanied by their reduced fecundity and enhanced catalase activity. Biochem Mol Biol Int 41:869–875. doi:10.1080/15216549700201911 334 J. Voller et al.

Skoog F, Strong FM, Miller CO (1965) Cytokinins. Science 148:532–533. doi:10.1126/science. 148.3669.532-a Slaugenhaupt SA, Mull J, Leyne M, Cuajungco MP, Gill SP, Hims MM, Quintero F, Axelrod FB, Gusella JF (2004) Rescue of a human mRNA splicing defect by the plant cytokinin kinetin. Hum Mol Genet 13:429–436. doi:10.1093/hmg/ddh046 Spíchal L, Rakova NY, Riefler M, Mizuno T, Romanov GA, Strnad M, Schmülling T (2004) Two cytokinin receptors of Arabidopsis thaliana, CRE1/AHK4 and AHK3, differ in their ligand specificity in a bacterial assay. Plant Cell Physiol 45:1299–1305. doi:10.1093/pcp/pch132 Spinola M, Colombo F, Falvella FS, Dragani TA (2007) N6-isopentenyladenosine: a potential therapeutic agent for a variety of epithelial cancers. Int J Cancer 120:2744–2748. doi:10.1002/ ijc.22601 Štarha P, Trávníček Z, Herchel R, Popa I, Suchý P, Vančo J (2009) Dinuclear copper(II) complexes containing 6-(benzylamino)purines as bridging ligands: synthesis, characterization, and in vitro and in vivo antioxidant activities. J Inorg Biochem 103:432–440. doi:10.1016/j. jinorgbio.2008.12.009 Strnad M (1997) The aromatic cytokinins. Physiol Plant 101:674–688. doi:10.1111/j.1399-3054. 1997.tb01052.x Tiedemann RE, Mao X, Shi C-X, Zhu YX, Palmer SE, Sebag M, Marler R, Chesi M, Fonseca R, Bergsagel PL, Schimmer AD, Stewart AK (2008) Identification of kinetin riboside as a repressor of CCND1 and CCND2 with preclinical antimyeloma activity. J Clin Invest 118:1750–1764. doi:10.1172/JCI34149 Tournas JA, Lin F-H, Burch JA, Selim MA, Monteiro-Riviere NA, Zielinski JE, Pinnell SR (2006) Ubiquinone, idebenone, and kinetin provide ineffective photoprotection to skin when compared to a topical antioxidant combination of vitamins C and E with ferulic acid. J Invest Dermatol 126:1185–1187. doi:10.1038/sj.jid.5700232 Tremaine AM, Ortiz A, Elkeeb L, Tran M, Weinstein G (2010) Long-term efficacy and safety of topical PRK 124 (0.125%) lotion (Pyratine-XR) in the treatment of mild-to-moderate rosacea. J drugs dermatology JDD 9:647–650 van der Hoeven D, Gizewski ET, Auchampach JA (2010) Activation of the A3 adenosine receptor inhibits fMLP-induced Rac activation in mouse bone marrow neutrophils. Biochem Pharmacol 79:1667–1673. doi:10.1016/j.bcp.2010.02.002 Verbeke P, Siboska GE, Clark BF, Rattan SI (2000) Kinetin inhibits protein oxidation and glycoxidation in vitro. Biochem Biophys Res Commun 276:1265–1270. doi:10.1006/bbrc. 2000.3616 Vermeulen K, Strnad M, Kryštof V, Havlíček L, Van der Aa A, Lenjou M, Nijs G, Rodrigus I, Stockman B, van Onckelen H, Van Bockstaele DR, Berneman ZN (2002) Antiproliferative effect of plant cytokinin analogues with an inhibitory activity on cyclin-dependent kinases. Leuk Off J Leuk Soc Am Leuk Res Fund UK 16:299–305. doi:10.1038/sj.leu.2402378 Veselý J, Havlíček L, Strnad M, Blow JJ, Donella-Deana A, Pinna L, Letham DS, Kato J, Detivaud L, Leclerc S (1994) Inhibition of cyclin-dependent kinases by purine analogues. Eur J Biochem 224:771–786. doi:10.1111/j.1432-1033.1994.00771.x Vičanová J, Bouez C, Lacroix S, Lindmark L, Damour O (2006) Epidermal and dermal characteristics in skin equivalent after systemic and topical application of skin care ingredients. Ann NY Acad Sci 1067:337–342. doi:10.1196/annals.1354.046 Voller J, Zatloukal M, Lenobel R, Doležal K, Béres T, Kryštof V, Spíchal L, Niemann P, Džubák P, Hajdúch M, Strnad M (2010) Anticancer activity of natural cytokinins: a structure-activity relationship study. Phytochemistry 71:1350–1359. doi:10.1016/j.phytochem.2010.04.018 Voller J, Béres T, Zatloukal M, Kaminski PA, Niemann P, Doležal K, Džubák P, Hajdúch M, Strnad M (2017) The natural cytokinin 2OH3MeOBAR induces cell death by a mechanism that is different from that of the “classical” cytokinin ribosides. Phytochemistry. doi:10.1016/j. phytochem.2017.01.004 14 Plant Hormone Cytokinins for Modulating Human Aging … 335

Walla J, Szücová L, Císařová I, Gucký T, Zatloukal M, Doležal K, Greplová J, Massino FJ, Strnad M (2010) X-ray structure, NMR and stability-in-solution study of 6-(furfurylamino)-9- (tetrahydropyran-2-yl)purine—a new active compound for cosmetology. J Mol Struct 975(1–3):376–380 Wang L, Sun C, Wang Z-H, Guo G-Q (2012) Mechanism of apoptotosis induced by ortho-topolin riboside in human hepatoma cell line SMMC-7721. Food Chem Toxicol 50:1962–1968. doi:10.1016/j.fct.2012.03.053 Wanitphakdeedecha R, Meeprathom W, Manuskiatti W (2015) Efﬁcacy and safety of 0.1% kinetin cream in the treatment of photoaging skin. Indian J Dermatol Venereol Leprol 81:547. doi:10.4103/0378-6323.157446 Werner T, Schmülling T (2009) Cytokinin action in plant development. Curr Opin Plant Biol 12 (5):527–538. doi:10.1016/j.pbi.2009.07.002 Wu JJ, Weinstein GD, Kricorian GJ, Kormeili T, McCullough JL (2007) Topical kinetin 0.1% lotion for improving the signs and symptoms of rosacea. Clin Exp Dermatol 32:693–695. doi:10.1111/j.1365-2230.2007.02513.x Yang B, Ji C, Kang J, Chen W, Bi Z, Wan Y (2009) Trans-zeatin inhibits UVB-induced matrix metalloproteinase-1 expression via MAP kinase signaling in human skin ﬁbroblasts. Int J Mol Med 23:555–560 Zhang K, Novak O, Wei Z, Gou M, Zhang X, Yu Y, Yang H, Cai Y, Strnad M, Liu C-J (2014) Arabidopsis ABCG14 protein controls the acropetal translocation of root-synthesized cytokinins. Nat Commun 5:3274. doi:10.1038/ncomms4274 Zwack PJ, Rashotte AM (2013) Cytokinin inhibition of leaf senescence. Plant Signal Behav 8: e24737. doi:10.4161/psb.24737 Palacký University in Olomouc Faculty of Science Laboratory of Growth Regulators & Department of Botany

A study of the effects of phytohormones on aging of

Caenorhabditis elegans

Summary of the Doctoral Thesis

Mgr. Alena Kadlecová

Study Programme: P1527 / Biology Field of Study: 1507V004 / Botany Form: Full-time

Olomouc 2019 Supervisor: Prof. Ing. Miroslav Strnad, CSc. DSc. Consultant: Mgr. Jiří Voller, Ph.D.

This PhD. thesis was carried out in the Laboratory of Growth Regulators within the framework of internal Ph.D. Study of Botany, guaranteed by the Department of Botany, Faculty of Science, Palacký University in Olomouc, between the years 2014-2019.

Ph.D. candidate: Mgr. Alena Kadlecová

Supervisor: Prof. Ing. Miroslav Strnad, DSc.

Laboratory of Growth Regulators, Centre of the Region Haná for Biotechnological and Agricultural Research, Palacký University & Institute of Experimental Botany ASCR, Olomouc, Czech Republic.

Opponents: Prof. Suresh Rattan, Ph.D., Dr.scient

Department of Molecular Biology and Genetics, Aarhus University, Aarthus, Denmark. Prof. RNDr. Hana Kolářová, CSc.

Department of Medicinal Biophysics, Faculty of Medicine and dentistry, Palacký University in Olomouc, Czech Republic.

MUDr. Petr Müller, Ph.D.

Department of Research – RECAMO, Masaryk Memorial Cancer Institute, Brno, Czech Republic.

The evaluation of this Ph.D. thesis was written by Prof. Ing. Aleš Lebeda, DrSc., Department of Botany, Faculty of Science, Palacký University in Olomouc.

The oral defence will take place on 13.12.2019 before the Commission for the Ph.D. thesis of the Study Program Botany, room SE-504, Šlechtitelů 27, Olomouc – Holice.

The PhD. thesis and expert reviews will be available 14 days before the defence in the Study Department of Faculty of Science (Mgr. M. Karásková), Palacký University, 17. listopadu 12, Olomouc.

After the defense, the Ph.D. thesis will be stored in the Library of the Biological Departments of Faculty of Science, Palacky University, Šlechtitelů 27, Olomouc – Holice.

Prof. Ing. Aleš Lebeda, DrSc. Chairman of the Commision for the Ph.D. thesis of Study Program Botany, Faculty of Science Palacký University in Olomouc

Contents

1. Introduction 4

2. Objectives 5

3. Materials and methods 6

4. Results and discussion 8

5. Conclusions and future perspectives 13

6. List of published papers and other contributions 14

7. Souhrn (in Czech) 16

References 16

1 Introduction

Cytokinins are a group of phytohormones implicated in various aspects of plant growth and development. Naturally occurring cytokinins are derivatives of adenine with either isoprenoid or aromatic side chain at position N6-. They stimulate division of variety of plant cells and regulate cell cycle progression, promote differentiation in some tissues, delay leaf senescence, promote chloroplast biogenesis, regulate circadian clock, influence nutrient uptake, modulate response to abiotic stress etc. [1, 2]. Various activities of cytokinins in animal models and even humans were also reported over the years. They were reported to have, for example, an antioxidative effect [3, 4, 5, 6], anti-aging activity in skin cells [7, 8] and flies [9, 6] or neuroprotective effect in several in vitro and in vivo models and even in humans [10, 11, 12, 13, 14, 15]. Some cytokinins underwent dermatological clinical trials and are currently used in cosmetics. Compounds with protective and pro-longevity effect could hopefully be used not only in cosmetics, but also medicine, and provide relief for elderly patients suffering from age- related pathologies such as neurodegenerative, cardiovascular and metabolic diseases. Cytokinins could be one group of such compounds. For their examination effect, we used the well established and convenient model organism Caenorhabditis elegans (C. elegans). C. elegans is a small, free-living soil nematode which can typically be found in temperate regions in rotting plant material, where it feeds on bacteria. It is safe, cheap and easy to maintain in a lab. It has a rapid life cycle and adults are able to produce hundreds of progeny within a week [16]. Aging worms also undergo many quantifiable changes equivalent to human aging, such as decrease in movement, sarcopenia, decline in sensory perception, reduced metabolic rates or accumulation or pigment lipofuscin [17, 18]. C. elegans is an eukaryote, therefore a significant percentage of its genes is homologous to the mammalian ones. These include many disease-relevant genes in humans [19]. The same is true for many molecular pathways that are conserved through the phylla. Due to these similarities, C. elegans is frequently used to study genetic and molecular aspects of aging and various human age-related diseases and can also be used for drug screenings [16, 20].

2 Objectives

The purpose of this work was to investigate the protective and pro-longevity effect of natural cytokinin bases and their derivatives using C. elegans. We studied the effect of natural cytokinin bases on the lifespan and stress resistance of the worms and performed initial experiments aimed at establishing the mechanism of action and understanding the metabolism of the best known active natural compound, kinetin. We also tested dozens of cytokinin derivatives, and found several that are potentially more active than their natural counterparts. It has been more than a decade since the topic was last comprehensively reviewed. This is why we also performed a literature survey, in which we summarized health- promoting activity of cytokinins, especially kinetin, in animal models.

3 Materials and methods

Strains

The bacterial strain used in this study was Escherichia coli OP50 (E. coli). C. elegans strains used in this study were: N2 (wild-type Bristol), BA17 (fem-1) and CF1038 (daf- 16(mu86)). All strains were purchased from Caenorhabditis genetic center (http://cbs.umn.edu/cgc/home).

Compounds

Natural cytokinin bases used in this study – kinetin (K), N6-benzylaminopurine (BAP), orthotopolin (oT), meta-topolin (mT), para-topolin (pT), N6-isopentenyladenine (iP), trans-zeatin (tZ), dihydrozeatin (DHZ), cis-zeatin (cZ) – were kindly provided by Olchemim s.r.o. Synthetic derivatives tested as a part of this thesis were prepared and provided by colleagues from Laboratory of Growth Regulators (LGR), Lenka Zahájská, Martin Hönig and Václav Mik. Prior to the testing, all compounds were disolved in dimethylsulfoxide (DMSO) to achieve 100 mM stock solutions. These were then added to cultivation media and therefore further diluted to reach the desired concentrations (up to 200 µM for the natural cytokinins, up to 100 µM for the derivatives).

Basic cultivation protocols

Worms were cultivated using standard protocols, buffers and media described in a peer- reviewed online resource WormBook (http://www.wormbook.org/, [21]). Maintenance cultures were kept in 15◦C. Experimental cultures were cultivated in 20◦C, or, in case of BA17 (fem-1), in 25◦C. In this temperature, the mutants develop into females and are unable to reproduce. To avoid contamination with progeny in experiments in which other strains were used, 25 µg/ml of fluorodeoxyuridine (FUDR) was added to the media. Chitinase assay

The short-term toxicity of natural cytokinin bases to the larvae and their effect on worm fecundity was measured via a chitinase assay [22] as well as by visual evaluation of the worms under a microscope.

Visual evaluation of lifespan in 96-well plates

The lifespan of experimental populations cultivated in liquid medium in 96-well plates was measured by manual counting of living worms in an inverted microscope [23]. Log- Rank p values were calculated from Kaplan-Meier survival curves.

Stress assays

Visual evaluation of stress resistance in 96-well plates

For part of the study, the resistance of experimental populations cultivated in liquid medium in 96-well plates to oxidative/heat stress was measured by visual counting of surviving worms in an inverted microscope. Log-Rank p values were calculated from Kaplan-Meier survival curves (oxidative stress assays) and Z test for population proportions was used to establish the statistical significance of the results from heat stress assays.

Automated microscopy and image analysis

Some experiments were performed using automatic microscopy and image analysis. This process required specialized equipment not available in the LGR at the moment, which is why these experiments were conducted during 3-months internship in Marta Artal Sanz’s group in Andalusian Centre for Developmental Biology, Universidad Pablo de Olavide, Sevilla. For microscopy, image analysis and parts of the high-throughput screening, we used protocols previously established in the hosting laboratory and published in Hernando-Rodríguez et al., 2018 [24], otherwise the protocol was similar to the one optimized in LGR. Results were analyzed using scripts in programming language Python (available at github), that were kindly created for us by Mgr. Jan Michelfeit. The method was used for preliminary evaluation of protective activity of cytokinin derivatives.

Preparation of worm extracts and cytokinin analysis by UHPLC-MS/MS

Adult worms and bacteria were cultivated with 100 µM K for 24h. The biological material was then washed 4 times, snap frozen with liquid nitrogen and stored in -80◦C. The extract preparation and UHPLC-MS/MS measurement of the levels of kinetin and its metabolites was performed by Hana Martínková and Dr. Ondˇrej Novák.

4 Results and discussion

Publication 1 Kadlecová, A., Jirsa, T., Novák, O., Kammenga, J., Strnad, M., & Voller, J. (2018). Natural plant hormones cytokinins increase stress resistance and longevity of Caenorhabditis elegans. Biogerontology, 19(2), 109- 120.

We tested the pro-longevity effect of 9 natural cytokinin bases. The set consisted of 5 cytokinins with aromatic side chain (kinetin (K), N6-benzylaminopurine (BAP), ortho- topolin (oT), meta-topolin (mT) and para-topolin (pT)) and 4 cytokinins with isporenoid substitution (N6-isopentenyladenine (iP), trans-zeatin (tZ), dihydrozeatin (DHZ) and cis- zeatin (cZ)). None of the compounds had acute toxic effect in C. elegans, nor did they significantly reduce fecundity of worms in the chitinase assay. In the initial screening, we identified four compounds that were able to significantly prolong the lifespan of the worms. Three of them were aromatic cytokinins – K, pT and mT. The remaining active cytokinin was tZ, however, we could not confirm its effect in repeated experiments with bigger populations. Commonly, interventions that prolong the lifespan of worms also increase their stress resistance [25]. In accordance with this trend, 3 day pre-treatment with K increased worms’ thermotolerance and resistance to superoxide anion generator [26] juglone. Next, we decided to perform initial steps to study the mechanism of action of K in C. elegans. The beneficial effect of K we observed could be explained by its interaction with one of the central longevity-associated pathways, insulin/insulin-like growth factor pathway (IIS). The main effector of this pathway is the transcription factor DAF- 16/FOXO (Forkhead box protein O) [27]. This underlay our decision to test the effect of K in daf-16(mu86) mutants with dysfunctional DAF-16 and we observed that K increased the stress resistance in these mutants as well. It therefore seems that K’s effect is not DAF- 16 dependent. The protective effect of cytokinins in animal models was previously ascribed to their ability to protect against oxidative stress. Reactive oxygen species (ROS) play an important role in aging. Traditionally, ROS were considered to be a crucial pro-aging factor because they cause a direct damage to cell’s macromolecules (free radical theory of

8 aging [28]). Today, we know that their role is much more complex – they can for example act as a signaling molecules, directly modify multiple proteins involved in signal transduction and they can also influence epigenetic modulation of gene expression [29]. Apart from this, ROS can also activate adaptive responses – this process is also known as mitochondrial hormesis or mitohormesis [30]. As K was previously proposed to act as hormetin, we decided to test how the presence of an antioxidant – analogue of vitamin E, Trolox – would influence the effect of K. Notably, in these co-treatment experiments the effect of K was significantly diminished in both stress assays and lifespan experiments. These results suggest that the mechanism of K’s action in worms cannot be ascribed to simply acting as a direct antioxidant – on the contrary, it seems that the presence of ROS is required for the beneficial effect of K. Our results are consistent with the hypothesis that K acts as a hormetin. Further, we conducted experiments aimed to determine how efficiently is K absorbed and how is it metabolized. We measured the levels of K in extracts prepared from non- treated and K-treated worms, and to assess how the bacteria used as food influence the metabolism, we also tested extracts from the bacteria and worms fed with heat-killed bacteria. No K was detected in samples from non-treated worms and bacteria. This suggests that, at least under the cultivation conditions we used, K is not naturally formed in these organisms. This finding is rather unexpected, as K was previously suggested to be formed in variety of different organisms as a result of DNA damage [31, 32, 33]. The published data concerning the natural occurrence of K is currently rather conflicted, as it was detected in native samples in some studies [34, 35, 36, 37], but not in other [38]. We also did not detect K in plant and cell extracts in some of our own experiments (unpublished). Exogenously applied K was efficiently absorbed by the worms and was also found in bacteria. Apart from K, we also detected kinetin riboside (KR) and kinetin riboside-5’- monophosphate (KRMP). The analytic method we used, did not allow us to detect the higher phosphates. It is therefore unclear whether also KRDP and KRTP are formed in the worms. All 3 metabolites were also found in extracts from worms fed with heat-killed bacteria, showing that C. elegans can effectively ribosylate K. In bacteria, KR but not KRMP was detected. The presence of these metabolites presents an interesting possibility that K might act as pro-hormetin and it is actually its metabolites that cause the hormetic response. KR is

9 well known cytotoxic compound able to induce ATP depletion [39, 40], possibly by impairing mitochondrial functions [41]. Low doses of KR might therefore trigger an adaptive response in C. elegans. Moreover, low doses of some cytokinin ribosides were previously reported to activate nuclear factor erythroid-derived 2-like 2 (Nrf2) [42] – a transcription factor playing an important role in regulating the stress response [43] – and some of our previously unpublished data suggest KR might have the same effect as well. Interestingly, K too was shown to activate Nrf2 in mouse hipoccampal cell line [14]. Taking all of our results together, there might be one more interesting possible explanation of K’s beneficial effect – K or its metabolites might act as an activator of AMPK. Both K and KR were previously reported to activate AMPK in cells [44] and one other cytokinin, N6-isopentenyladenosine-5’monophosphate, was proposed to activate AMPK by binding to its γ-subunit instead of AMP [45]. It is tempting to speculate that KRMP, into which K/KR can be transformed to in worms and cells [46], can do that as well. It is known that the effect of AMPK can be suppressed by addition of antioxidants [47, 48, 49], as we saw in case of K. The effect of AMPK on aging of worms is also likely at least partly independent on DAF16, as the C. elegans daf-16 and aak-2 (gene coding AMPK in worms) double mutants have shorter lifespan than either single mutants [50]. Moreover, metformin, an antidiabetic drug able to prolong the lifespan of multiple species most likely via AMPK activation [51], also prolongs the lifespan of daf-16 mutants [52].

Publication 2 Kadlecová, A., Maková, B., Artal-Sanz, M., Strnad, M. & Voller, J. (2019). The plant hormone kinetin in disease therapy and healthy aging. Ageing Research Reviews, in press.

The protective activity of natural cytokinins prompted us to investigate their synthetic derivatives. Over more than 20 years of existence of LGR, hundreds of cytokinin derivatives were prepared for application in medicine, cosmetics and agriculture, resulting in a unique chemical library. One notable example of cytokinin derivative with protective activity created by LGR is 6-furfurylamino-9-(tetrahydropyran-2-yl)-9H-purine, a derivative of K which is currently used in cosmetics and marketed under the trade name Pyratine. The close link between the increased lifespan and stress resistance proves to be useful for drug screenings for compounds modulating aging, as stress assays are more suitable for high-throughput screening (HTS) methods. This is why we, in cooperation with Dr. Marta ArtalSanz’s group from Universidad Pablo de Olavide, decided to conduct HTS for compounds increasing resistance of C. elegans to oxidative and heat stress. Our test set included derivatives of natural cytokinin bases with substitutions at various positions of the purine ring. The tested set was enriched for compounds with protective activity, with 24 compounds out of 73 being able to significantly increase resistance to one of both types of stresses in the initial screening (Z-test for population proportions). The majority of the hits have a substitution at position 8 of the purine ring and some of them also possess 9- (tetrahydropyran-2-yl) group, same as the approved cosmaceutical Pyratine. The preparation of these compounds was described in detail here [53]. Natural cytokinins were also included in the set. Notably, many of the active compounds were more active than their natural counterparts, partly maybe because here, the compounds were tested at a lower concentration than in the study focused on the protective activity of natural cytokinin bases. What makes these results even more promising is that we have recently shown that some these derivatives also have a protective effect in models of mitochondrial disease (unpublished). Together with the fact that they show only marginal toxicity to human non-cancer cell lines [53], we believe they might be an interesting candidates for both drug and cosmetics development.

Publication 3

Hönig, M., Plíhalová, L., Spíchal, L., Grúz, J., Kadlecová, A., Voller, J., Svobodová, A.R., Vostálová, J., Ulrichová, J., Doležal, K., & Strnad, M. (2018). New cytokinin derivatives possess UVA and UVB photoprotective effect on human skin cells and prevent oxidative stress. European journal of medicinal chemistry, 150, 946-957.

Another example of active compounds we discovered during our investigation of cytokinin derivatives are aromatic cytokinins with 9-(tetrahydrofuran-2-yl) moiety. 13 compounds were prepared and tested. We found that two of them, namely 6-(thiophen-2-yl methylamino)9-(tetrahydrofuran-2-yl)purine and 2-chloro-6-furfurylamino-9- (tetrahydrofuran-2-yl)purine, as well as 6-furfurylamino-9-tetrahydrofuran-2-yl (Kin-THF), were able to significantly increase the stress resistance of worms. Their protective effect was not limited to a single model, as these compounds also protected human skin cells against UVA and UVB irradiation, and they were non-toxic in various human skin cell lines. Although their mechanism of action has not yet been investigated, the lack of activity in ORAC assay suggests that the derivatives do not act as direct antioxidants.

5 Conclusions and future perspectives

We have established that C. elegans is a suitable model for investigating protective and pro-longevity effect of cytokinins. We discovered that 3 natural aromatic cytokinin bases – kinetin (K), orthotopolin and para-topolin – prolong the lifespan of C. elegans. K, which was selected as a representative compound for follow-up experiments, also increased thermotolerance and resistance of worms to oxidative stress. It was efficiently absorbed and metabolized by worms into kinetin riboside and kinetin riboside-5’- monophosphate. We also showed that the effect of K was DAF-16 independent, but, interestingly, the presence of ROS seems to be required for its activity. Although these results do not allow us to find a definite answer to the question of how K induces longevity and stress resistance in worms, they permit us to form several plausible hypotheses and provided basis for follow-up study, in which the involvement of SKN-1 (worm homolog of Nrf2) and AMPK will be investigated. The absence of K in native worm and bacterial samples also prompted us to conduct another follow-up study, in which the presence and metabolism of cytokinins in various species under normal and stress conditions will be examined. We also screened for active synthetic derivatives of natural cytokinin bases using both traditional and modern automated high-throughput screening methods. In our experiments, derivatives with substitution in position 8 of the purine ring and aromatic cytokinins with 9(tetrahydrofuran-2-yl) moiety seemed to be especially promising. Their protective effect will be further examined in the future. Cytokinins are interesting candidates for medicine and dermatology, and since the 1990s, when their beneficial activity in animal models was described for the first time, dozens of papers regarding their protective effect have been published. However, it has been more than 15 years since the topic was last comprehensively reviewed and these older works understandably do not contain more recent exciting discoveries, such as the effect of K on alternative splicing of pre-mRNA [10] or the ability of KRTP to act as a neo-substrate of disease relevant kinases [12, 13]. This motivated us to summarize current knowledge in a book chapter and a review article focused on kinetin.

6 List of published papers and other contributions

Papers in journals with impact factor

1. Kadlecová, A., Jirsa, T., Novák, O., Kammenga, J., Strnad, M., & Voller, J. (2018). Natural plant hormones cytokinins increase stress resistance and longevity of Caenorhabditis elegans. Biogerontology, 19(2), 109-120.

2. Kadlecová, A., Maková, B., Artal-Sanz, M., Strnad, M. & Voller, J. (2019). The plant hormone kinetin in disease therapy and healthy aging.Ageing Research Reviews, in press.

Book chapter

Conference contributions

2. Kadlecová, A. Protective effect of cytokinins and their derivatives in Caenorhabditis elegans / News from the „worm cabinet“. CBPRS 2018, Luhačovice 2018. Oral presentation. 3. Kadlecová, A., Jirsa, T., Strnad, M., Voller, J. Cytoprotective activity of natural cytokinin bases in Caenorhabditis elegans. Trends in Natural Product Research –

PSE Young Scientists’ Meeting, Natural Products in Health, Agro-Food and Cosmetics, Lille 2017. Poster.

5. Kadlecová, A. Worms getting high-throughput. Growth regulators on the way, Malá Morávka 2016. Oral presentation.

6. Voller, J., Kadlecová, A. Databáze látek zpomalujících stárnutí. Národní bioinformatická konference ENBIK, Loučeň, 2016. Poster.

8. Voller, J., Plíhalová, L., Zahájská, L., Zatloukal, M., Kadlecová, A., Grúz, J., Nardelliová, B., Schubert, D., Spíchal, L., Strnad, M. Anti-aging activity of phytohormones cytokinins. Healthy Aging: From Molecules to Organisms, Hinxton 2015. Poster.

9. Kadlecová, A. Studying cytoprotective activity of cytokinins in Caenorhabditis elegans. Biotechnology of Phytohormones and Natural Substances, Malá Morávka 2015. Oral presentation.

Awarded grants

1. INTER-COST no. LTC17071 – Hodnocení účinků rostlinných hormonů a jejich derivátů na háďátko obecné. Principal investigator: Mgr. Alena Kadlecová. Duration: 2017 – 2019.

7 Souhrn (in Czech)

Název dizertační práce: Studium vlivu fytohormonů na stárnutí Caenorhabditis elegans

Cytokininy, fytohormony podílející se na růstu a vývoji rostlin, jsou také známy pro své různorodé účinky v živočiších. V této práci jsme se zaměřili na protektivní účinky cytokininových bází. Ty jsou dle dřívějších studií schopny například zpomalovat stárnutí lidských kožních buněk či octomilek a mají antioxidativní a neuroprotektivní efekt v in vitro i in vivo modelech. O jejich účincích jsme publikovali souhrnný článek a kapitolu v knize. Pro podrobnější studium jejich aktivity jsme zvolili háďátko obecné (Caenorhabditis elegans), jakožto dobře zavedený model stárnutí. Zjistili jsme, že 3 přírodní aromatické cytokininy – kinetin, para-topolin a meta-topolin – dokáží prodloužit délku života háďátek. Kinetin (K), nejlépe prostudovaný cytokinin s protektivními účinky, byl vybrán jako reprezentativní látka pro další experimenty. UHPLC-MS/MS experimenty odhalily, že K je háďátky dobře vstřebáván a přeměňován na kinetin ribosid a kinetin ribosid-5'- monofosfát. Kinetin dokázal také zvýšit rezistenci C. elegans k oxidativnímu a teplotnímu stresu. Jeho efekt nebyl závislý na DAF-16, centrálnímu efektoru jedné z nejznámějších drah, jež dokáže regulovat stárnutí, tzv. "insulin/insulin-line growth factor signalling pathway". Přítomnost antioxidantu nicméně dokázala příznivý efekt kinetinu potlačit. To naznačuje, že K v C. elegans nepůsobí jednoduše jako přímý antioxidant, ale že přítomnost kyslíkových radikálů je naopak nezbytná pro jeho efekt. Toto zjištění je konzistentní s dřívější hypotézou, že K, či případně některý z jeho metabolitů, působí jako hormetin. V plánovaných navazujících studiích bude prozkoumán efekt kinetinu na SKN- 1, funkční C. elegans ortolog savčího Nrf2 proteinu, a na AMP-aktivovanou protein kinázu. Předpokládá se, že malé množství K přirozeně vzniká v široké škále organizmů při poškození DNA. Z tohoto důvodu je zajímavé, že jsme K nedetekovali v nativních vzorcích z C. elegans a bakterií. Proto také plánujeme další studii, kde bude podrobněji prozkoumána přítomnost a metabolismus K i jiných cytokininů v řadě modelových organismů, a to jak za normálních tak za stresových podmínek. Kromě přírodních látek jsme také prozkoumali efekt několika desítek syntetických cytokininových derivátů. Jakožto neslibnější se v našich experimentech jevily deriváty se substitucí v pozici 8 purinového kruhu a aromatické cytokininy s 9-(tetrahydrofuran-2-yl) skupinou. I tyto látky budou v budoucnu podrobněji prostudovány.

References

[1] J. J. Kieber and G. E. Schaller, “Cytokinins,” The Arabidopsis Book/American Society of Plant Biologists, vol. 12, 2014.

[2] T. Werner and T. Schmülling, “Cytokinin action in plant development,” Current Opinion in Plant Biology, vol. 12, no. 5, pp. 527–538, 2009.

[3] A. Olsen, G. E. Siboska, B. F. Clark, and S. I. Rattan, “N6-furfuryladenine, kinetin, protects against Fenton reaction-mediated oxidative damage to DNA,” Biochemical and Biophysical Research Communications, vol. 265, no. 2, pp. 499–502, 1999.

[4] P. Verbeke, G. E. Siboska, B. F. Clark, and S. I. Rattan, “Kinetin inhibits protein oxidation and glycoxidation in vitro,” Biochemical and Biophysical Research Communications, vol. 276, no. 3, pp. 1265–1270, 2000.

[5] A. Brizzolari, C. Marinello, M. Carini, E. Santaniello, and P. A. Biondi, “Evaluation of the antioxidant activity and capacity of some natural N6-substituted adenine derivatives (cytokinins) by fluorimetric and spectrophotometric assays,” Journal of Chromatography B, vol. 1019, pp. 164–168, 2016.

[6] S. P. Sharma, J. Kaur, and S. I. Rattan, “Increased longevity of kinetin-fed Zaprionus fruitflies is accompanied by their reduced fecundity and enhanced catalase activity,” IUBMB Life, vol. 41, no. 5, pp. 869–875, 1997.

[7] S. I. Rattan and B. F. Clark, “Kinetin delays the onset of aging characteristics in human fibroblasts,” Biochemical and biophysical research communications, vol. 201, no. 2, pp. 665–672, 1994.

[8] S. I. Rattan and L. Sodagam, “Gerontomodulatory and youth-preserving effects of zeatin on human skin fibroblasts undergoing aging in vitro,” Rejuvenation Research, vol. 8, no. 1, pp. 46–57, 2005.

[9] S. P. Sharma, P. Kaur, and S. I. Rattan, “Plant growth hormone kinetin delays aging, prolongs the lifespan, and slows down development of the fruitfly Zaprionus paravittiger,” Biochemical and Biophysical Research Communications, vol. 216, no. 3, pp. 1067–1071, 1995.

[10] S. A. Slaugenhaupt, J. Mull, M. Leyne, M. P. Cuajungco, S. P. Gill, M. M. Hims, F. Quintero, F. B. Axelrod, and J. F. Gusella, “Rescue of a human mRNA splicing defect by the plant cytokinin kinetin,” Human Molecular Genetics, vol. 13, no. 4, pp. 429–436, 2004.

[11] G. Gold-von Simson, J. D. Goldberg, L. M. Rolnitzky, J. Mull, M. Leyne, A. Voustianiouk, S. A. Slaugenhaupt, and F. B. Axelrod, “Kinetin in familial dysautonomia carriers: implications for a new therapeutic strategy targeting mRNA splicing,” Pediatric Research, vol. 65, no. 3, p. 341, 2009.

[12] N. T. Hertz, A. Berthet, M. L. Sos, K. S. Thorn, A. L. Burlingame, K. Nakamura, and K. M. Shokat, “A neo-substrate that amplifies catalytic activity of Parkinson’s-disease-related kinase PINK1,” Cell, vol. 154, no. 4, pp. 737–747, 2013.

[13] L. E. Bowie, T. Maiuri, M. Alpaugh, M. Gabriel, N. Arbez, D. Galleguillos, C. L. Hung, S. Patel, J. Xia, N. T. Hertz, et al., “N6-furfuryladenine is protective in Huntington’s disease models by signaling huntingtin phosphorylation,” Proceedings of the National Academy of Sciences, vol. 115, no. 30, pp. E7081–E7090, 2018.

[14] Y. Wei, D. Liu, Y. Zheng, C. Hao, H. Li, and W. Ouyang, “Neuroprotective effects of kinetin against glutamate-induced oxidative cytotoxicity in HT22 cells: Involvement of Nrf2 and Heme Oxygenase- 1,” Neurotoxicity Research, vol. 33, no. 4, pp. 725–737, 2018.

[15] S. J. Choi, C.-H. Jeong, S.-G. Choi, J.-Y. Chun, Y. J. Kim, J. Lee, D.-H. Shin, and H. J. Heo, “Zeatin prevents amyloid β-induced neurotoxicity and scopolamine-induced cognitive deficits,” Journal of Medicinal Food, vol. 12, no. 2, pp. 271–277, 2009.

[16] A. Barrière and M.-A. Félix, “Isolation of C. elegans and related nematodes,” WormBook, vol. 17, pp. 1–9, 2006.

[17] H. A. Tissenbaum, “Using C. elegans for aging research,” Invertebrate Reproduction & Development, vol. 59, no. sup1, pp. 59–63, 2015.

[18] J. J. Collins, C. Huang, S. Hughes, and K. Kornfeld, “The measurement and analysis of age-related changes in Caenorhabditis elegans,” WormBook, 2007.

[19] E. Culetto and D. B. Sattelle, “A role for Caenorhabditis elegans in understanding the function and interactions of human disease genes,” Human Molecular Genetics, vol. 9, no. 6, pp. 869–877, 2000.

[20] T. Kaletta and M. O. Hengartner, “Finding function in novel targets: C. elegans as a model organism,” Nature Reviews Drug Discovery, vol. 5, no. 5, p. 387, 2006.

[21] T. Stiernagle, “Maintenance of C. elegans,” C. elegans, vol. 2, pp. 51–67, 1999.

[22] B. R. Ellerbrock, E. M. Coscarelli, M. E. Gurney, and T. G. Geary, “Screening for presenilin inhibitors using the free-living nematode, Caenorhabditis elegans,” Journal of Biomolecular Screening, vol. 9, no. 2, pp. 147–152, 2004.

[23] G. Solis and M. Petrascheck, “Measuring Caenorhabditis elegans life span in 96 well microtiter plates,” Journal of Visualized Experiments: JoVE, no. 49, 2011.

[24] B. Hernando-Rodríguez, A. P. Erinjeri, M. J. Rodríguez-Palero, V. Millar, S. González-Hernández, M. Olmedo, B. Schulze, R. Baumeister, M. J. Muñoz, P. Askjaer, et al., “Combined flow cytometry and high-throughput image analysis for the study of essential genes in Caenorhabditis elegans,” BMC Biology, vol. 16, no. 1, p. 36, 2018.

[25] K. Zhou, Z. Pincus, and F. Slack, “Longevity and stress in Caenorhabditis elegans,” Aging (Albany NY), vol. 3, pp. 733–753, Aug 2011.

[26] J. J. Inbaraj and C. F. Chignell, “Cytotoxic action of juglone and plumbagin: a mechanistic study using HaCaT keratinocytes,” Chemical Research in Toxicology, vol. 17, no. 1, pp. 55–62, 2004.

[27] C. Kenyon, J. Chang, E. Gensch, A. Rudner, R. Tabtiang, et al., “A C. elegans mutant that lives twice as long as wild type,” Nature, vol. 366, no. 6454, pp. 461–464, 1993.

[28] D. Harman, “Aging: a theory based on free radical and radiation chemistry,” Journal of Gerontology, vol. 11, no. 3, pp. 298–300, 1955.

[29] P. Davalli, T. Mitic, A. Caporali, A. Lauriola, and D. D’Arca, “ROS, cell senescence, and novel molecular mechanisms in aging and age-related diseases,” Oxidative Medicine and Cellular Longevity, vol. 2016, pp. 1– 18, 2016.

[30] M. Ristow and K. Schmeisser, “Mitohormesis: promoting health and lifespan by increased levels of reactive oxygen species (ROS),” Dose-Response, vol. 12, no. 2, pp. dose–response, 2014.

[31] J. Barciszewski, G. E. Siboska, B. O. Pedersen, B. F. Clark, and S. I. Rattan, “A mechanism for the in vivo formation of N6-furfuryladenine, kinetin, as a secondary oxidative damage product of DNA,” FEBS letters, vol. 414, no. 2, pp. 457–460, 1997.

[32] J. Barciszewski, G. E. Siboska, B. O. Pedersen, B. F. Clark, and S. I. Rattan, “Furfural, a precursor of the cytokinin hormone kinetin, and base propenals are formed by hydroxyl radical damage of DNA,” Biochemical and Biophysical Research Communications, vol. 238, no. 2, pp. 317–319, 1997.

[33] E. Wyszko, M. Z. Barciszewska, M. Markiewicz, M. Szymanski, W. T. Markiewicz, B. F. Clark, and J. Bar-´ ciszewski, ““Action-at-a distance” of a new DNA oxidative damage product 6-furfuryladenine (kinetin) on template properties of modified DNA,” Biochimica et Biophysica Acta (BBA)- Gene Structure and Expression, vol. 1625, no. 3, pp. 239–245, 2003.

[34] J. Barciszewski, G. E. Siboska, B. O. Pedersen, B. F. Clark, and S. I. Rattan, “Evidence for the presence of kinetin in DNA and cell extracts,” FEBS letters, vol. 393, no. 2-3, pp. 197–200, 1996.

[35] J. Barciszewski, M. Mielcarek, M. Stobiecki, G. Siboska, and B. F. Clark, “Identification of 6- furfuryladenine (kinetin) in human urine,” Biochemical and Biophysical Research Communications, vol. 279, no. 1, pp. 69– 73, 2000.

[36] R. S. Shetty, C. S. Gallagher, Y.-T. Chen, M. M. Hims, J. Mull, M. Leyne, J. Pickel, D. Kwok, and S. A. Slaugenhaupt, “Specific correction of a splice defect in brain by nutritional supplementation,” Human Molecular Genetics, vol. 20, no. 21, pp. 4093–4101, 2011.

[37] L. Ge, J. W. H. Yong, N. K. Goh, L. S. Chia, S. N. Tan, and E. S. Ong, “Identification of kinetin and kinetin riboside in coconut (Cocos nucifera L.) water using a combined approach of liquid chromatography– tandem mass spectrometry, high performance liquid chromatography and capillary electrophoresis,” Journal of Chromatography B, vol. 829, no. 1-2, pp. 26–34, 2005.

[38] A. L. Orr, F. U. Rutaganira, D. de Roulet, E. J. Huang, N. T. Hertz, K. M. Shokat, and K. Nakamura,

“Long-term oral kinetin does not protect against α-synuclein-induced neurodegeneration in rodent models of Parkinson’s disease,” Neurochemistry International, vol. 109, pp. 106–116, 2017.

[39] Y. Ishii, Y. Hori, S. Sakai, and Y. Honma, “Control of differentiation and apoptosis of human myeloid leukemia cells by cytokinins and cytokinin nucleosides, plant redifferentiation-inducing hormones,” Cell Growth and Differentiation, vol. 13, no. 1, pp. 19–26, 2002.

[40] C. Cabello, W. Bair, S. Ley, S. Lamore, S. Azimian, and G. Wondrak, “The experimental chemotherapeutic N6-furfuryladenosine (kinetin-riboside) induces rapid ATP depletion, genotoxic stress, and CDKN1A (p21) upregulation in human cancer cell lines,” Biochemical Pharmacology, vol. 77, no. 7, pp. 1125–1138, 2009.

[41] J. Cheong, D. Goh, J. Yong, S. Tan, and E. Ong, “Inhibitory effect of kinetin riboside in human heptamoa, HepG2,” Molecular BioSystems, vol. 5, no. 1, pp. 91–98, 2009.

[42] A. Dassano, M. Mancuso, P. Giardullo, L. De Cecco, P. Ciuffreda, E. Santaniello, A. Saran, T. A. Dragani, and F. Colombo, “N6-isopentenyladenosine and analogs activate the NRF2-mediated antioxidant response,” Redox Biology, vol. 2, pp. 580–589, 2014.

[43] Q. Ma, “Role of Nrf2 in oxidative stress and toxicity,” Annual Review of Pharmacology and Toxicology, vol. 53, pp. 401–426, 2013.

[44] D. Mijikovic, J. Hranisavljevic, and Z. Pietrzkowski, “Pharmaceutical compositions and methods for metabolic modulation,” Aug. 5 2004. US Patent App. 10/567,875.

[45] S. Pisanti, P. Picardi, E. Ciaglia, L. Margarucci, R. Ronca, A. Giacomini, A. M. Malfitano, A. Casapullo, C. Laezza, P. Gazzerro, et al., “Antiangiogenic effects of N6-isopentenyladenosine, an endogenous isoprenoid end product, mediated by AMPK activation,” The FASEB Journal, vol. 28, no. 3, pp. 1132–1144, 2014.

[46] P. Mlejnek and P. Doležel, “Apoptosis induced by N6-substituted derivatives of adenosine is related to intracellular accumulation of corresponding mononucleotides in HL-60 cells,” Toxicology in vitro, vol. 19, no. 7, pp. 985–990, 2005.

[47] T. Schulz, K. Zarse, A. Voigt, N. Urban, M. Birringer, and M. Ristow, “Glucose restriction extends Caenorhabditis elegans life span by inducing mitochondrial respiration and increasing oxidative stress,” Cell Metabolism, vol. 6, no. 4, pp. 280–293, 2007.

[48] P. Mungai, G. Waypa, A. Jairaman, M. Prakriya, D. Dokic, M. Ball, and P. Schumacker, “Hypoxia triggers AMPK activation through reactive oxygen species-mediated activation of calcium release- activated calcium channels,” Molecular and Cellular Biology, vol. 31, no. 17, pp. 3531–3545, 2011.

[49] W. De Haes, L. Frooninckx, R. Van Assche, A. Smolders, G. Depuydt, J. Billen, B. Braeckman, L. Schoofs, and L. Temmerman, “Metformin promotes lifespan through mitohormesis via the

peroxiredoxin PRDX-2,” Proceedings of the National Academy of Sciences, vol. 111, no. 24, pp. E2501–E2509, 2014.

[50] J. Apfeld, G. O’Connor, T. McDonagh, P. S. DiStefano, and R. Curtis, “The AMP-activated protein kinase AAK-2 links energy levels and insulin-like signals to lifespan in C. elegans,” Genes & Development, vol. 18, no. 24, pp. 3004–3009, 2004.

[51] M. G. Novelle, A. Ali, C. Diéguez, M. Bernier, and R. de Cabo, “Metformin: a hopeful promise in aging research,” Cold Spring Harbor Perspectives in Medicine, vol. 6, no. 3, p. a025932, 2016.

[52] B. Onken and M. Driscoll, “Metformin induces a dietary restriction–like state and the oxidative stress response to extend C. elegans healthspan via AMPK, LKB1, and SKN-1,” PloS One, vol. 5, no. 1, p. e8758, 2010.

[53] L. Zahajská, J. Nisler, J. Voller, T. Gucký, T. Pospíšil, L. Spíchal, and M. Strnad, “Preparation, characterization and biological activity of C8-substituted cytokinins,” Phytochemistry, vol. 135, pp. 115–127, 2017.