Bioassay-Guided Isolation of Antiproliferative Compounds from Limbarda crithmoides (L.) Dumort.

Limbarda crithmoides (L.) Dumort (Asteraceae) n-hexane extract displayed high cell proliferation inhibitory activity against acute myeloid leukaemia cells (OCI-AML3) and was therefore subjected to a bioassay-guided multistep separation procedure. Two thymol derivatives, namely 10-acetoxy-8,9-epoxythymol tiglate (1) and 10-acetoxy-9-chloro-8,9-dehydrothymol (2), were isolated and identified by means of NMR spectroscopy. Both of them exhibited a significant dose-dependent inhibition of cell proliferation.


Introduction
The genus Limbarda (Asteraceae), formerly included in the genus Inula, comprises two accepted species: L. crithmoides (L.) Dumort., and L. salsoloides (Turcz.) Ikonn [1]. L. crithmoides is a halophyte plant, commonly known as Inula crithmoides (L.), that is widespread across the salt marshes and sea cliffs of the Mediterranean Sea, French Atlantic coasts, English channel and western European seaboards [1]. L. crithmoides is a common edible plant. Its leaves, eaten raw or boiled, are a good source of protein, amino acids, fibers, vitamins and other components, and traditionally formed an important part of the Lebanese diet [2]. Young shoots of L. crithmoides are also consumed as pickles in vinegar in Spain [3], while its raw tops are added to salads in the Basilicata region of southern Italy [4]. I. crithmoides biological activities have been reviewed, along with other eight Asteraceae representative species [5]. Essential oil composition has been widely explored [6][7][8], and it showed to be dependent on the region where plants grow [8].
Extracts of L. crithmoides have been applied to crops and weeds to investigate their properties, which confirmed their herbicidal potency [9]. It was also reported that the extracts of callus cultures show good antifungal activity against Alternaria solani and Phytophthora cryptogea [10]. In a previous work, methanol and hexane extracts of the aerial parts of this plant were found to reduce the radial growth of A. solani and P. cryptogea, and they also show weak antifungal activity against fungi of the Fusarium genus [11]. The antioxidant activity of L. crithmoides has been widely investigated [11][12][13][14], and seems to be directly correlated with the presence of phenolic metabolites such as quinic acid derivatives [12,[15][16][17]. For example, the ethyl acetate fraction of a methanolic extract of I. crithmoides showed in vitro and in vivo hepatoprotective activity against carbon tetrachloride (CCl 4 )-induced liver injury through antiradical and antioxidant activities [18]. In addition to antifungal activity, antimicrobial as well as antileishmanial activities of L. crithmoides extracts have also been reported [15,[19][20][21].
As part of our ongoing research into antiproliferative compounds derived from natural sources [22][23][24][25][26], a phytochemical investigation of L. crithmoides was undertaken to isolate active compounds by means of a bioassay-guided fractionation of active extracts against the Ontario Cancer Institute-Acute Myeloid Leukemia-3 (OCI-AML3) acute myeloid leukaemia cell line.
In a preliminary study, 24 h of treatment with 100 and 200 µg/mL of the methanolic extract (M) of L. crithmoides induced a significant decrease in OCI-AML3 cell number ( Figure S1A Figure S4B shows a representative experiment of the experiment series quantified in Figure S4A. Thus, the H and DCM extracts were at least 10-fold more active than the M extract ( Figures S3 and S4, Supplementary Material), and the H extract was more potent than the DCM extract in increasing apoptotic cell death. For this reason, we decided to continue only with the H extract.
The serial bioassay-guided chromatographic procedures led to the isolation of two active compounds (1 and 2; Figure 1). Using NMR spectroscopic data, compound 1 was identified as 10-acetoxy-8,9-epoxythymol tiglate, previously isolated from Athrixia spp. [43], Schkuhria multiflora Hook & Arn. [44] and Eupatorium cannabinum L. [45], and compound 2 was characterized as 10-acetoxy-9Z-chloro-8,9-dehydrothymol, previously isolated from Arnica sachalinensis (Regel) A. Gray [46] (Figures S5-S8, Supplementary Material). The Z configuration of the double bond was confirmed by NOESY experiment ( Figure S9). Thymol derivatives are quite common in different Inula spp. [47][48][49], and it is known that L. crithmoides produces chlorinated thymol derivatives [37]. However, as far as we know, this is the first time that compounds 1 and 2 have been isolated from L. crithmoides (I. crithmoides). The two pure compounds were tested for their biological activity.              Thus, compounds 1 and 2 were very effective in decreasing the number of OCI-AML3 cells through both increasing apoptosis and blocking cell proliferation. Compound 2 was more potent than compound 1 as it exerted inhibitory effects at a concentration as low as 1.25 µg/mL. The increased expression of p21 is largely associated with the arrest of cell cycle by two different pathways. The first is a p53-dependent and the second is p-53 -independent pathway. This prompted us to examine the expression of both p21 and p53 in OCI-AML3 cells untreated or treated with the M, DCM, H extracts and the compounds 1 and 2 isolated from the latter as a possible mechanism of cell cycle arrest since these pathways regulates mitotic progression and promotes cellular stress response [50]. We used Western blotting to measure expression of p21 in control and treated cells. As shown in Figure 6, upper panel, all the tested extracts and compounds induced an upregulation of p21, but only with the H extract and with compound 2 the increase of p21 was significant compared to controls ( Figure 6, lower left panel). Because p21 is regulated through either p53-dependent or -independent pathways, we also measured expression of p53, which did not change after treatment with any of the extracts or compounds (Figure 6, upper and lower right panels). Thus, compounds 1 and 2 were very effective in decreasing the number of OCI-AML3 cells through both increasing apoptosis and blocking cell proliferation. Compound 2 was more potent than compound 1 as it exerted inhibitory effects at a concentration as low as 1.25 µg/mL. The increased expression of p21 is largely associated with the arrest of cell cycle by two different pathways. The first is a p53-dependent and the second is p-53 -independent pathway. This prompted us to examine the expression of both p21 and p53 in OCI-AML3 cells untreated or treated with the M, DCM, H extracts and the compounds 1 and 2 isolated from the latter as a possible mechanism of cell cycle arrest since these pathways regulates mitotic progression and promotes cellular stress response [50]. We used Western blotting to measure expression of p21 in control and treated cells. As shown in Figure 6, upper panel, all the tested extracts and compounds induced an upregulation of p21, but only with the H extract and with compound 2 the increase of p21 was significant compared to controls ( Figure 6, lower left panel). Because p21 is regulated through either p53-dependent or -independent pathways, we also measured expression of p53, which did not change after treatment with any of the extracts or compounds (Figure 6, upper and lower right panels). These results suggest that the H extract induced a significant activation of a p21-dependent, p53-independent pathway in OCI-AML3 cells and that compound 2 was responsible of this activity and the associated cell cycle arrest. Notably, the p21-dependent, p-53-independent pathway has been shown to be associated with an increased apoptosis, as is evident in our system, in contrast to the p21-dependent, p53-dependent pathway that promotes the translocation of p21 from the nucleus to the cytoplasm thus determining an anti-apoptotic effect absent in our study [22].
It is known from previous studies that compounds isolated from Inula species may have antiproliferative effect on a different leukaemia cells [51][52][53][54][55][56][57][58][59] but, as far as we know, only one work describing the activity of Inula extract on an acute myeloid leukaemia (KG1a cell line) has been reported [60]. This work highlighted the effect of Inula compounds on induction of apoptosis by the mitochondria-dependent pathway. Most of the Inula metabolites tested were sesquiterpenolides. On the other hand, thymol and thymol derivatives as well as extracts and essential oils containing these compounds [61][62][63][64] are known to be cytotoxic. These results suggest that the H extract induced a significant activation of a p21-dependent, p53-independent pathway in OCI-AML3 cells and that compound 2 was responsible of this activity and the associated cell cycle arrest. Notably, the p21-dependent, p-53-independent pathway has been shown to be associated with an increased apoptosis, as is evident in our system, in contrast to the p21-dependent, p53-dependent pathway that promotes the translocation of p21 from the nucleus to the cytoplasm thus determining an anti-apoptotic effect absent in our study [22].
It is known from previous studies that compounds isolated from Inula species may have antiproliferative effect on a different leukaemia cells [51][52][53][54][55][56][57][58][59] but, as far as we know, only one work describing the activity of Inula extract on an acute myeloid leukaemia (KG1a cell line) has been reported [60]. This work highlighted the effect of Inula compounds on induction of apoptosis by the mitochondria-dependent pathway. Most of the Inula metabolites tested were sesquiterpenolides. On the other hand, thymol and thymol derivatives as well as extracts and essential oils containing these compounds [61][62][63][64] are known to be cytotoxic.

Plant Material
Aerial parts of L. crithmoides were collected during the flowering period (August 2016) in Fano, Urbino, Italy. Specimen collection was restricted to coastal habitats, nearly always within the reach of sea spray. Specimens were authenticated by Dr Laura Giamperi (University of Urbino). Voucher specimens have been deposited at the Herbarium of the Botanic Garden of the University of Urbino (GS 203).

Cell Line Culture and Characterisation
The antiproliferative activity of the two compounds under consideration was tested on the acute myeloid leukaemia (AML) cell line OCI-AML3 [65]. OCI-AML3 cells were maintained in Roswell Park Memorial Institute (RPMI) 1640 medium with 10% foetal bovine serum (FBS), 100 U/mL penicillin, and 100 µg/mL streptomycin at 37 • C and 5% CO 2 . Cells were purchased from ATCC (LGC Standards S.r.l., Sesto San Giovanni, Milan, Italy), kept at logarithmic growth and cultured in 24-well plates to assess their number and morphology. Cultures kept at 2 × 10 5 cells/mL were treated with different concentrations of DMSO or the test compounds at the final concentrations reported in the figures. These reported concentrations were chosen based on preliminary experiments. After 24 h, the cell number was quantified using a haemocytometer.

Analysis of Cell Viability and Cell Cycle Progression
Cell viability and cell cycle progression were examined by flow cytometry to measure the amount of DNA in nuclei stained with propidium iodide (PI; Sigma-Aldrich, Milan, Italy), with the exclusion of necrotic cells by forward light scatter (FSC) [66]. Briefly, cells were harvested by centrifugation and gently resuspended in 1.5 mL hypotonic PI solution (50 µg/mL in 0.1% sodium citrate plus 0.1% Triton X-100). Tubes were kept in the dark at 4 • C for 30 min. PI fluorescence of individual nuclei was measured by flow cytometry using a Coulter Epics XL-MCLTM flow cytometer (Beckman Coulter, Cassina De' Pecchi, Milan, Italy) and analysed using FlowJo_V10 software (BD Biosciences, Milan, Italy).

Conclusions
This study showed that the M extract of L. crithmoides has cell proliferation inhibitory activity against acute myeloid leukaemia cells (OCI-AML3) and solvent partition showed that the H fraction was more active than the DCM fraction. For this reason, we decided to continue testing the H fraction. Chromatographic purification of fraction H led to the isolation of two active thymol derivatives, 1 and 2. Compound 2 was shown to be highly active. As far as we know, this is the first report of the antiproliferative activity of chlorinated thymol derivatives. The DCM fraction, to a minor extent, was shown to still be active, and for this reason we think it would be interesting to explore its composition and the antiproliferative activity of its fractions and compounds.