Cattleianal and Cattleianone: Two New Meroterpenoids from Psidium cattleianum Leaves and Their Selective Antiproliferative Action against Human Carcinoma Cells

The Myrteacae family is known as a rich source of phloroglucinols, a group of secondary metabolites with notable biological activities. Leaves of Psidium cattleianum were extracted with chloroform: methanol 8:2 to target the isolation of phloroglucinol derivatives. Isolated compounds were characterized using different spectroscopic methods: nuclear magnetic resonance (NMR), ultra-violet (UV) and mass spectrometry (MS). Two new phloroglucinols were evaluated for cytotoxicity against a panel of six human cancer cell lines, namely colorectal adenocarcinoma cells (HT-29 and HCT-116); hepatocellular carcinoma cells (HepG-2); laryngeal carcinoma (Hep-2); breast adenocarcinoma cells (MCF7 and MDA-MB231), in addition to normal human melanocytes HFB-4. Additionally, cell cycle analysis and annexin-V/FITC-staining were used to gain insights into the mechanism of action of the isolated compounds. The new phloroglucinol meroterpenoids, designated cattleianal and cattleianone, showed selective antiproliferative action against HT-29 cells with IC50’s of 35.2 and 32.1 μM, respectively. Results obtained using cell cycle analysis and annexin-V/FITC-staining implicated both necrosis and apoptosis pathways in the selective cytotoxicity of cattleianal and cattleianone. Our findings suggest that both compounds are selective antiproliferative agents and support further mechanistic studies for phloroglucinol meroterpenoids as scaffolds for developing new selective chemotherapeutic agents.


Introduction
Myrtaceous plants exert a myriad of biological activities including anti-diabetic, cytotoxic, anti-inflammatory and anti-bacterial biofilm activities [1][2][3]. Psidium cattleianum Sabine (Family Myrtaceae) is a shrub or small tree native to Brazil, where it is known as "araçá" [4]. It is now cultivated throughout the tropics and subtropics for its juicy purplered fruits know as strawberry guava or Cattley guava [5]. In a recent report, ninety-three compounds were identified in the aqueous leaves extract of P. cattleianum where catechin, gallic acid and vanillic acid were reported as the major constituents [6].
Leaves of P. cattleianum are used in folk medicine as an anti-hemorrhagic, antispasmodic and anti-diarrheal agent [7], but recent investigations have uncovered a selective anti-proliferative effect of P. cattleianum leaf extract against human carcinoma cells [8,9].
Additionally, a closely related species, Psidium guavaja, showed cytotoxic activity in prostate cancer cell lines that was mediated through the induction of apoptosis [10]. Phytochemical investigation of P. guajava resulted in the isolation of sixteen phloroglucinol meroterpenoids which all displayed cytotoxicity against five human cell lines with IC 50 reported below 60 µM [11]. Among isolated meroterpenoids, psiguajavadial A and psiguajavadial B induced apoptosis in human cancer cell line HCT116 in the nano-molar range. Most recently, six novel phloroglucinol meroterpenoids were isolated from Psidium littorale Raddi which all showed cytotoxic activity against several human cancer cell lines [12,13] emphasizing the potential of phloroglucinol meroterpenoids as versatile and selective cytotoxic agents. Therefore, this study was designed to target the isolation of novel phloroglucinols from P. cattleianum and investigate their possible anti-proliferative activity.

Identification of the Isolated Compounds
Two new meroterpenoids were isolated from the chloroform: methanol extract of Psidium cattleianum leaf. Structure elucidation of the isolated compounds was carried out based on their m/z, recorded using high-resolution mass spectrometry HR-MS, UV absorbance and 1D and 2D-NMR spectra.
Another EI-MS fragment at m/z 203 was consistent with the sesquiterpene moiety of βcaryophyllene as previously reported for littordial A-E, guajadial A and pisidial A [12,16,17]. 13 C-NMR and 1 H-13 C-HSQC established the caryophyllene moiety encompassing three methyls, six methylenes, three methine and three quaternary carbons. The presence of a pyran ring connecting the phloroglucinol and sesquiterpene moiety was inferred from quaternary carbon (C-4) at δ C 87.02 and its HMBC correlation to the phloroglucinol (H-9 2H, 2.88, 3.04), which in turn showed clear HMBC correction to the aromatic carbon C-1 at δ C 104.7 ppm, Figure 1, supplementary material Figures S4-S7. To complete the assignment of the sesquiterpene moiety, an exocylic double bond was evidenced by the presence of methylene group (C-15), observed in the HSQC spectrum at δ H/C 4.84,4.86/110.1, Figures S5 and S6. Also, a cyclobutane ring was determined by the characteristic chemical shift for C-1 at δ H/C (1.38, m/58.3), characteristic of other phloroglucinol meroterpenoids psidial A and littordial B and C [11][12][13][14]. Further analysis of HMBC and HSQC spectra of compound 1 and careful comparison of the observed chemical shifts of C-1, C-3 and C-6 with that of the chemical literature of other phloroglucinols suggested that compound 1 is most structurally related to littordial B and C which were previously isolated from P. littorale [12] and pisidial A [11], previously isolated from P. guajava which all share 1R,4R,5S,9S stereochemistry, as can be inferred from the chemical shifts of C-1, C-3 and C-6 [15,18]. Accordingly, compound 1 was identified as a β-caryophyllene-phloroglucinol derivative and, based on the presence of two aldehyde groups, was designated as cattleianal, Figure 1.  Figure S10). Another significant difference was the appearance of two methyl groups at δ H 1.99 (s, 3H) and 0.89 (t, J = 6.7, 3H). 13 C-NMR showed the characteristic signals of β-caryophyllene moiety and confirmed lack of formyl substitution, (Supplementary Data, Figure S11). Instead, 13 C-spectrum showed a signal for a quaternary carbon at 205.1 ppm consistent with an aromatic ketone and an upfield methyl group at δ C 6.02 with clear HMBC correlations to δ C 161.5, 156.4, Supplementary Data, Figures S12 and S13. This indicated the position of the methyl substituent (C-7 ) between two aromatic carbons with hydroxyl groups C-4 and C-6 , respectively. Also, the unusual upfield chemical shift in the aromatic methyl can be explained by its m-position to the ketone substituent at δ C 205.1.
Meanwhile, the carbonyl of the ketone group was traced via HMBC correlation to a C3 aliphatic chain (C-9 -C-11 ), in addition to the correlations observed in the COSY spectrum (Figures S16 and S17), establishing the presence of 3 -butanoyl, 5 -methyl-benzene moiety instead of 3 , 5 -diformyl benzyl commonly seen in Psidium meroterpenoids [14,15]. To the best of our knowledge, littrodial D is the only meroterpenoid with β-caryophyllene moiety and ketone substituent of the phloroglucinols identified so far [12,18]. However, no derivative has been isolated in which the other formyl group is replaced with a methyl substituent as observed in compound 2. Accordingly, compound 2 was designated as cattleianone, a new β-caryophyllene meroterpenoid with 3 -butanoyl, 5 -methylphloroglucinol.

Cytotoxicity Assessment of Cattleianal and Cattleianone
Cytotoxicity evaluation of cattleianal and cattleianone against a panel of human cell lines indicated that both compounds have weak cytotoxic activity against HCT-116 cells with calculated IC 50 's above 70 µM. However, HepG2 and HT-29 were more susceptible to the antiproliferative/cytotoxic effects of both compounds, Table 1. This highlights the selectivity of both compounds, in which IC 50 's could not be determined against normal human melanocytes (HFB4 cells) up to concentration > 100 µg/mL.

Apoptosis/Necrosis Assessment Using Flow Cytometry
Plant-derived natural products can often induce cell death by either apoptosis or necrosis, while some, such as resveratrol, are reported to display both pathways [19]. We assessed cell fractions going through apoptosis/necrosis after incubation with both drugs and observed that both compounds induced only necrotic non-programmed cell death in HCT-116 cells, with no significant change in early or late apoptotic cell fractions after 24 h incubation ( Figure 3A).
Meanwhile, in HT-29 cells, cattleianal and cattleianone, at their IC 50 concentration, induced a significant increase in the number of cells undergoing apoptosis (early and late) as well as cells undergoing non-specific necrosis compared to untreated cells ( Figure 3B). In MDA-MB-231, only cattleianone induced significant apoptosis while both compounds induced non-programmed cell death (necrosis) when compared to control cells as shown in Figure 3C.

Molecular Assessment of Apoptosis
Further molecular confirmation of apoptosis was undertaken by assessing the expression level of several key apoptosis related genes using qPCR technique. Both cattleianal and cattleianone induced significant overexpression of the p53 gene by 2.  Figure 4B,C). This is not only indicative of p53 overexpressing the effect of cattleianal and cattleianone, but also indicative of p53 activation effect of cattleianone in these two cell lines. None of the compounds under investigation induced any significant change in the expression of either p53 or TP53INP1 genes in HCT-116 cells ( Figure 4A).  In addition, the concentration of the effective apoptosis executional marker, active caspase-3, was determined after treatment with cattleianal and cattleianone using the ELISA technique. Caspase-3 concentration was significantly higher in response to treatment with cattleianal and cattleianone in HT-29 cells. However, only cattleianone induced a significant increase in the active caspase-3 concentration in MDA-MB-231 cells ( Figure 4D).

Discussion
Phloroglucinols are a class of phenolic secondary metabolites with wide distribution in the Myrtaceae family, but were also reported in other plant families as well as in microorganisms and marine sources [18]. Phloroglucinols can occur as monomers, dimers or as terpene adduct such as phloroglucinol meroterpenoids of Eucalyptus and Psidium [18]. Phloroglucinols of both genera have been reported to possess cytotoxic effects including euglobals and macrocarpals reported in Eucalyptus species and psidials, littrodials from Psidium species [11][12][13]20]. Herein, we report the isolation and cytotoxic activity of two new phloroglucinol meroterpenoids, namely cattleianal and cattleianone, from the leaves of P. cattelianum. Both compounds are phloroglucinol-β-caryophyllene adducts similar to those reported in other Psidium species [13,14]. Both cattleianal and cattleianone lack substituent at C-9 which set them apart from phloroglucinols of P. guajava (often substituted with a phenyl ring at C-9 ) and phloroglucinols from P. littorale, where C-9 position is often substituted with C3 and C4 aliphatic chain [13,14]. Preliminary investigations have indicated that several phloroglucinols meroterpenoids displayed cytotoxic activity, including some at the nanomolar range [13][14][15].
Previous reports have established the cytotoxicity of P. cattleianum leaves' extract [10,11], and our findings here point to contributions of cattleianal and cattleianone to this reported cytotoxicity. Cattleianal and cattleianone showed somewhat selective but weak to moderate activity against the investigated cancer cell lines where the cytotoxic activity was more pronounced in Hep-2 and HT-29 cell lines. Interestingly, previous investigation of the cytotoxic potential of P. cattleianum leaf extract found similar differential toxicity against cancer cell lines, with human gastric carcinoma (SNU-16) reported as a highly susceptible cell line [8].
The results of cell cycle distribution analysis of cattleianal and cattleianone on HCT-116, HT29 and MDA-MB231 cell lines suggested that the antiproliferative effect of both compounds at their IC 50 is likely mediated through G 0 /G 1 cycle arrest. In a previous study, other phloroglucinol derivatives; namely sideroxylonal-B and Macrocarpal-A, showed a similar anti-proliferating effect against human breast carcinoma cells (MCF7), by reducing percentage of cells in the S-phase through increasing percentage of cells in G 0 /G 1 phase [20].
Further analysis using Annexin V-FITC/PI flow cytometry indicated that both compounds induced early and late apoptosis in HT-29 cells, implicating apoptosis related proteins. Supporting our results, Caspases and PARP proteins were previously reported to be involved in the cytotoxicity of P. cattelianum extract against gastric carcinoma cells [8]. Additionally, natural and semisynthetic phloroglucinols have been shown to activate Caspases 3 and 9 with subsequent cleavage of PARP protein in human glioma and breast carcinoma cell lines [16,17]. Therefore, it is likely that Caspases are one of the target proteins mediating the antiproliferative action of cattleianal and cattleianone. Yet, to further examine this hypothesis, p53 and TP53INP1 gene expression were assessed and found to be induced by cattleianal and cattleianone treatment. The gene p53 carries a central role in apoptosis and the TP53INP1 gene is responsible for phosphorylation and activation of p53-induced apoptosis signaling [21,22]. The elevated concentration of the active form of caspase-3 in response to cattleianal and cattleianone treatment confirms the propagation of apoptosis signal to the late apoptotic phase in affected cell lines [23] Meanwhile, HCT-116 cells, which were more resistant to both compounds, underwent necrosis showed no significant cell population in early or late apoptosis. A similar pattern was observed for MDA-MB231 breast carcinoma cells, which were more resistant to cattleianal and went through necrosis after incubation with the drug (IC 50 64.4 µM), while upon incubation with cattleianone (IC 50 38.7 µM), MDA-MB231 cells underwent apoptosis. Thus, these results implicate both programmed and non-programmed cell death in the cytotoxicity effect of cattleianal and cattleianone. It can be argued that the differential activity of both compounds against different cell lines is likely related to the expression levels of their apoptosis-effector targets in each cell line where lower availability of the target protein(s) makes the cells more resistant to the effect of the drugs, and necrosis pathway is triggered as a result. Further investigations into the specific target(s) are therefore recommended for a better understanding of the chemotherapeutic benefits for both drugs.

Extraction and Isolation
One kilogram of powdered leaves was extracted with chloroform: methanol 80:20 in a Soxhlet apparatus following the method described by Sidana et al. [24] for the targeted extraction of phloroglucinols, to yield 114 g of dark residue after solvent evaporation. The extract (62 g) was chromatographed on a VLC column (10 × 15 cm) of silica gel H (330 g). Elution sequence started with n-hexane (100%) followed by 5% increments of EtOAc, up to 100% EtOAc, then CHCl 3 /MeOH (5% increments), up to 20% MeOH. Fractions (200 mL each) were collected and monitored by TLC and fractions with similar TLC patterns were pooled together to yield 4 collective major fractions (Fr. I-Fr. IV).

Cytotoxicity Assays
Cytotoxicity was tested by SRB assay as previously described [25]. Briefly, exponentially growing cells were collected using 0.25% trypsin-EDTA and plated in 96-well plates at 1000-2000 cells/well. Cells were exposed to test compounds for 72 h and subsequently fixed with TCA (10%) for 1 h at 4 • C. After washing trice, cells were exposed to 0.4% SRB solution for 10 min in dark then washed with 1% glacial acetic acid. After drying overnight, Tris-HCl was used to dissolve SRB-stained cells and color intensity was measured at 540 nm. Doxorubicin ® (Sigma Company, Suffolk County, NY, USA) was used as a standard cytotoxic drug.
The dose response curve of each compound was analyzed using Emax model (Equation (1)): where [R] is the residual unaffected fraction (the resistance fraction), [D] is the drug concentration used, [Kd] or IC 50 is the drug concentration that produces a 50% reduction of the maximum inhibition rate and [m] is a Hill-type coefficient. Absolute IC 50 is defined as the drug concentration required to reduce absorbance by 50% of control (i.e., Kd = absolute IC 50 when R = 0 and E max =100-R) [26].

Cell Cycle Analysis
To assess the effect of tested compounds on cell cycle distribution, cells were treated with the pre-calculated IC 50 for each compound for 24 h. After treatment, cells were harvested with trypsin/EDTA, washed twice with ice-cold PBS and then re-suspended in 0.5 mL of PBS. Cells were fixed with 60% ice-cold ethanol for a minimum of one hour at 4 • C and stored at −20 • C. After two washes with PBS, cells were resuspended in 1 mL of PBS containing 50 µg/mL RNAase-A and 10 µg/mL propidium iodide (PI) and incubated for 20 min in dark at room temperature. DNA content was analyzed by ACEA Novocyte™ flowcytometer (ACEA Biosciences Inc., San Diego, CA, USA) and PI fluorescent signals were analyzed using FL2 detector (λex/em 535/617). For each sample, 12,000 events were acquired and cell cycle distribution was calculated using ACEA NovoExpress™ software (ACEA Biosciences Inc., San Diego, CA, USA) [27].

Apoptosis Assessment
To assess the effect of tested compounds on programmed cell death, early/late apoptosis and necrosis cell populations were measured using Annexin V-FITC/PI apoptosis/necrosis detection kit (Abcam Inc., Cambridge Science Park, Cambridge, UK). Briefly, cells under investigation were treated with the pre-determined IC 50 's for each compound for 24 h, then cells were collected by trypsinization, washed twice with ice-cold PBS, and resuspended in 0.5 mL of annexin V-FITC/PI solution for 30 min in the dark according to manufacturer protocol. After staining at room temperature, cells were injected through ACEA Novocyte™ flowcytometer (ACEA Biosciences Inc., San Diego, CA, USA) and analyzed for FITC and PI fluorescent signals using FL1 and FL2 signal detector, respectively (λex/em 488/530 nm for FITC and λex/em 535/617 nm for PI). For each sample, 12,000 events were acquired and positive FITC and/or PI cells were quantified by quadrant analysis and calculated using ACEA NovoExpress™ software (ACEA Biosciences Inc., San Diego, CA, USA) [28].

Apoptosis Gene Expression Analysis
Real-time polymerase chain reaction (PCR) was performed to assess the expression of p53 and TP53INP1 genes after treatment with the pre-determined IC 50 's of cattleianal and cattleianone. After 24h of treatment, RNA was extracted using mirVana™ RNA isolation kit (Invitrogen, Carlsbad, CA, USA). The RNA and purity were confirmed (A260/280 > 2.0) using DeNovix DS-11™ microvolume spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). Subsequently, the total RNA samples of all treatments were reverse transcribed to construct a cDNA library using the SuperScript™ Master Mix kit (Invitrogen, Carlsbad, CA, USA). The cDNAs were then subjected to quantitative real-time PCR reactions using TaqMan ® Gene Expression Assay (Applied Biosystems, Foster City, CA, USA). p53 forward primer was 3 -CCTCACCATCATCACACTGG-5 , and backward primer was 5 -CTGAGTCAGGCCCTTCTGTC-3 ; TP53INP1 forward primer was 3 -TTCCTCCAACCAAGAACCAGA-5 , and backward primer was 5 -GCTCAGTAGGTGACT CTTCACT-3 . Housekeeping β-actin gene was used as a reference gene with forward primer 3 -GAGAGGCGGCTAAGGTGTTT-5 and backward primer 5 -TGGTGTAGACGG GGATGACA-3 ; normalized fold changes for all genes of interest were calculated using the formula: 2 −∆∆Cq .

Assessment of Active Caspase-3 Concentration
To confirm the effect of cattleianal and cattleianone on apoptosis, the active caspase-3 level was measured using Quantikine ® caspase-3 ELISA kit (R&D Systems, Inc., Minneapolis, MN, USA). Briefly, after cattleianal and cattleianone treatment for 24 h, cells were collected and washed, then incubated with the biotin-ZVKD-fmk inhibitor for 1 h. Cells were transferred into the wells of a microplate pre-coated with an anti caspase-3 monoclonal antibody and washed to remove any unbound substances. Streptavidin conjugated to horseradish peroxidase was added to the wells and bound to the biotin on the inhibitor. Following a wash to remove any unbound Streptavidin-HRP, a substrate solution was added to the wells. The enzyme reaction yields a blue product that turned yellow when a Stop Solution was added. The optical density of each well was determined within 30 min, using a microplate reader set to 450 nm with a wavelength correction at 540 nm or 570 nm. The concentrations of active caspase-3 were calculated from a standard curve of, constructed with known concentrations of active caspase-3. Caspase concentration was expressed as ng/mg protein. Protein concentrations were determined using Bradford method and purified bovine serum albumin was used as a standard calibration protein.

Statistical Analysis
Data are presented as mean ± SEM using GraphPad prism™ software (GraphPad software Inc., La Jolla, CA, USA) for windows version 5.00. Analysis of variance (ANOVA) with Dunnett's post hoc test was used for testing the significance using SPSS ® for windows, version 17.0.0. p < 0.05 was taken as a cut off value for significance.

Conclusions
Phytochemical investigation of the leaves of P. cattleianum resulted in the isolation of two new meroterpenoid phloroglucinols (cattleianal and cattleianone) with selective cytotoxic activity against HT-29, Hep-2, MCF7 and MDA-MB231 cancer cell lines. In colorectal adenocarcinoma cells (HT-29), both compounds showed an anti-proliferative effect by causing G 0 /G 1 cell cycle arrest which was accompanied by induction of early and late apoptosis. Meanwhile, HCT-116 cells, the most resistant to both compounds, under-went only necrosis after 24 h of incubation implicating non-specific cell death. The study emphasizes previous work regarding the potential use of phloroglucinol derivatives as selective anti-proliferative/cytotoxic agents and encourages further research in this group of compounds as a scaffold for development of selective chemotherapeutic agents.