Plants in the genus Kalanchoe (Crassulaceae), though originating mostly in Madagascar and Southeast Africa, have a global distribution in warm climates. Frequently, Kalanchoe spp. occur as exotic or invasive species. Many members of the genus are able to self-propagate from plantlets produced on the leaf margin, making established populations hard to eradicate (Descoings, 2003; Akulova-Barlow, 2009). The presence of toxic cardiac glycosides make some Kalanchoe spp. a grazing hazard for animals in agriculture, with documented issues in Brazil, South Africa, and Australia (Botha C., 2013; Botha C.J., 2013; Mendonça et al., 2018). Nevertheless, these plants display a diverse array of stunning forms and are often grown as ornamentals for their strange beauty.

Despite their often exotic presence, Kalanchoe spp. have ethnobotanical uses wherever they are found, sometimes being called “miracle leaf” for their use in treating various ailments (Akulova-Barlow, 2009; Milad et al., 2014). In the developing world, members of this genus are used for treating myriad medical conditions. Because of its widespread distribution and ubiquitous ethnobotanical use, much research has been focused on K. pinnata, a species native to Madagascar but cultivated and distributed throughout the tropics (Descoings, 2003; Biswas et al., 2011a; Quazi Majaz et al., 2011; Pattewar, 2012; Rajsekhar et al., 2016). This species has even been the subject of bioengineering – a transgenic K. pinnata that produces an antimicrobial peptide (AMP cecropin P1) has recently been developed (Zakharchenko et al., 2016; Lebedeva et al., 2017).

Because the genus has demonstrated medicinal potential, Kalanchoe spp. neglected in research should be explored for bioactive compounds. K. mortagei and K. fedtschenkoi, two members of the section Bryophyllum within the genus, are two such species with established ethnobotanical usage, but which have been overlooked in natural products research.

Kalanchoe mortagei, also known by the synonyms K. poincarei or Bryophyllum mortagei, is a plant native to rocky/sandy soils in north Madagascar (Descoings, 2003). Compared to other members of the genus, little research has been conducted on the chemical and medicinal properties of this species (Maiti et al., 1995). Despite this, K. mortagei is grown in Mexican homegardens, and its leaves are taken orally for digestive disorders and as a local remedy for cancer in Antioquia Department, Colombia (Blanckaert et al., 2004; Vera-Marín and Sánchez-Sáen, 2016). The roots of the plant are used for treating parasitic worm-related diseases in parts of Indonesia (Herawati and Husin, 2000).

Kalanchoe fedtschenkoi is a perennial native to central/southern Madagascar but is naturalized well outside its original range (Descoings, 2003). Introduced populations can be found in Florida, Texas, and Puerto Rico (USDA/NRCS, 2013). A popular garden succulent, K. fedtschenkoi is a model organism for research into Crassulacean acid metabolism (CAM) (Dittrich, 1976; Nimmo et al., 1986; Cook et al., 1995). In Brazil, this species is used as an analgesic (Cumberbatch, 2011).

The rise of antimicrobial resistant (AMR) bacterial infections is one of the most pressing issues in medicine. Increasingly, conventional antibiotic medications are failing to stop persistent and dangerous bacterial diseases (Irenji et al., 2018; Katsuura et al., 2018). A report commissioned by the UK government notes that roughly 700,000 people die annually from AMR infections; this figure is projected to increase to 10 million deaths per year by 2050 (O’Neil, 2016) and encompasses data from across the broad spectrum of pathogenic microbes. In the face of rising morbidity and mortality due to AMR infections, the need for new drugs to address drug-resistance is clear (van der Meer et al., 2014). In 2015, the WHO launched the Global Antimicrobial Resistance Surveillance System (GLASS) to unify worldwide AMR. To date they have collected data from 42 countries and received over 500,000 AMR pathogenic strains (WHO, 2017).

Six bacterial species, the “ESKAPE” pathogens, have been highlighted by the Infectious Disease Society of America (IDSA) as being especially dangerous due to their patterns of antibiotic resistance. They are responsible for the majority of nosocomial infections worldwide (Table 1) (Boucher et al., 2009).

Plants used in traditional medicine are a potential source for novel antimicrobial compounds (Rahman et al., 2018; Salam and Quave, 2018). In the developing world, the large majority of people (75%) rely on plants for primary medical needs, including for wound healing and antimicrobial agents (Sarker et al., 2005). Historically, the bulk of manufactured drugs were derived from plant natural products, and the majority of these drugs were tied directly to their original ethnobotanical use (Chin et al., 2006; Sarker and Nahar, 2012). Even between 1982 and 2002, 79% of approved drugs worldwide had a natural product origin (Chin et al., 2006).

Secondary metabolites taken from plants used in traditional medicine have been found to inhibit microbial growth and virulence. Kalanchoe spp. have demonstrated such antimicrobial properties, and have been proven to accelerate wound-healing. For example, extracts and compounds from K. pinnata are effective against cutaneous leishmaniasis, a disease caused by trypanosome protozoa (Torres-Santos et al., 2003; Muzitano et al., 2006a,b, 2009).

In the past decade, substantial research has examined the antibacterial properties of K. pinnata and several other Kalanchoe spp. Extracts of K. blossfeldiana, K. crenata, K. laciniata, and K. pinnata have all demonstrated growth inhibitory effects on over 15 bacterial species, including four of the ESKAPE pathogens (Tables 2, 3).

In 15 studies that evaluated the antimicrobial effects of Kalanchoe spp., 12 focused solely on K. pinnata, one on K. laciniata (Iqbal et al., 2016) and one on K. blossfeldiana, a common household ornamental (Sarkar et al., 2015). A 2007 study compared the growth-inhibitory properties of K. crenata favorably with K. pinnata (Akinsulire et al., 2007).

Ten studies examined methanolic extracts, the most common solvent tested. Ethanol and water (five studies each) were also frequently used solvents. Research has established that methanolic crude extracts of K. pinnata outperform aqueous extracts in their growth-inhibitory effects (Akinsulire et al., 2007; Majaz et al., 2011; Nwadinigwe, 2011; Pattewar et al., 2013); this is also true for K. crenata (Akinsulire et al., 2007).

Studies also established the antibacterial effects of flavonoids extracted from K. pinnata (Okwu and Nnamdi, 2011; Tatsimo et al., 2012), as well as its leaf juice (Obaseiki-Ebor, 1985; Akinsulire et al., 2007). At least one study demonstrated the effects of K. pinnata in vivo, looking at how aqueous extracts accelerate the healing of wounds infected with Staphylococcus aureus and/or Pseudomonas aeruginosa (Lebedeva et al., 2017).

Research has firmly established K. pinnata as a plant of medicinal interest, and the overall genus continues to show promise as a potential source of antimicrobial, antibacterial compounds.

The aim of this study was to evaluate the antimicrobial potential of two previously neglected species: K. mortagei and K. fedtschenkoi against a panel of clinically relevant ESKAPE pathogens.

Two plant species were used in this experiment. Kalanchoe mortagei plants were grown from a specimen collected by the first author (NR) in Bradenton, FL, United States, in May 2008 (27.468591, -82.577127). A single K. fedtschenkoi plant was procured from the University of Georgia Plant Biology Greenhouse in Athens, GA, United States, in 2015. All plant material used in this experiment came from plants propagated from these two mother specimens. Plants were grown in NR’s personal collection and at the Emory University Greenhouse. Voucher specimens of each species were deposited at the Emory University Herbarium (GEO), and species identification confirmed by Dr. Tharanga Samarakoon at GEO (Accession nos.: 22702 and 22474 for K. fedtschenkoi and K. mortagei, respectively). Specimens were digitized and are available for viewing on the SERNEC portal (SERNEC, 2018).

Bulk plant materials were harvested, dried in a dehumidification chamber, and homogenized in a Waring blender into a fine powder. Retention vouchers of dried and ground material were prepared for future reference and stored in Quave Research Group laboratories at Emory University.

A total of seven crude extracts were prepared, four from K. mortagei and three from K. fedtschenkoi (Table 4). Each extract represented a particular plant part or combination of parts, though extract creation was also guided by limitations in available plant biomass.

Dry, ground plant biomass was double macerated for 72 h each with either 80 or 95% ethanol at a 1:10 ratio (w/v). The extracts were agitated daily and then vacuum filtered. The aqueous extract (1509aq) was prepared as a decoction; the dry plant material was boiled with deionized water (dH₂O) for 20 min and then filtered. After filtration the solvent was removed by rotary evaporation at ≤40°C. Extracts were redissolved in dH₂O, shell frozen in a dry ice-acetone bath, and then lyophilized overnight on a Labconco FreeZone 2.5 Lyophilizer (Kansas City, MO, United States). Dry extracts were scraped into scintillation vials and stored at -20°C. Organic extracts were dissolved in DMSO and the aqueous extract was re-dissolved in dH₂O to yield a stock concentration of 10 mg mL^-1 for microbiological assays.

Mammalian cytotoxicity of extracts was assessed using human keratinocytes (HaCaTs) and a lactate dehydrogenase (LDH) test kit (G-Biosciences, St. Louis, MO, United States) as previously described (Quave et al., 2015). Briefly, HaCaTs were maintained in Dulbecco’s modified Eagle’s medium with L-glutamine and glucose supplemented with 10% heat-inactivated fetal bovine serum and 1× solution of penicillin and streptomycin at 37°C, 5% CO₂ in 75 mL flasks. Once 90–95% confluency was reached, the cells were detached from the flask bottom using 0.25% trypsin, 0.1% ethylenediaminetetraacetic acid (EDTA) in Hanks’ balanced salt solution (HBSS) without Ca⁺⁺, Mg⁺⁺, and NaHCO₃. The culture was standardized to 4 × 10⁴ cells mL^-1 using a hemocytometer. Then, 200 μL of the standardized culture was added to each well in a 96-well tissue culture-treated microtiter plate (Falcon 35–3075) and the plates were incubated for 48 h in a humidified 37°C, 5% CO₂ incubator, prior to media aspiration. Either media containing extracts (4–512 μg mL^-1) or vehicle were serially diluted and processed 24 h later following manufacturer’s protocol for chemical induced cytotoxicity. Percent DMSO (v/v) in the wells was <2% for all tests.

Each extract was characterized by HPLC using a method adapted from four previously published HPLC methods, one examining flavonoid compounds (Nielsen et al., 2005), and three examining bufadienolides (a type of cardiac glycoside commonly found in Kalanchoe plants) (Supratman et al., 2000; Huang et al., 2013; Moniuszko-Szajwaj et al., 2016). Extracts were dissolved in methanol (1465, 1469), methanol:dH₂O (1420, 1421, 1509aq), or methanol:dH₂O:DMSO (1468, 1508). All extracts were chromatographed on an Agilent 1260 Infinity system running OpenLab CDS ChemStation (Agilent Technologies, Santa Clara, CA, United States) with an Agilent ZORBAX Eclipse XDB-C18 (250 mm × 4.6 mm, 5 μm) column with compatible guard column at 30°C. A 10 μL injection of each extract was eluted at a flow rate of 1 mL min^-1 using a mobile phase consisting of (A) 0.1% formic acid in H₂O and (B) 0.1% formic acid in methanol (VWR HiPerSolv CHROMANORM). The gradient profile consisted of initial conditions 98:2 A:B which were held for 20 min, then increased to 24.5:75.5 A:B from 20 to 95.5 min, and finally to 100% B at 110 min, which was held for 20 min. Chromatograms of each extract were generated using ultraviolet–visual spectroscopy (UV–vis) during HPLC, and reported at 254 nm.

Standard flavonoids, kaempferol (MP Biomedicals, Inc.), and quercetin (Enzo Life Sciences), as well as phenolic compounds, caffeic acid, p-coumaric acid, and ferulic acid (MP Biomedicals, Inc.) were used to aid in characterization by HPLC.

Total phenolic content (TPC) was determined using a Folin–Ciocalteu assay modified for 96-well plate format (Singleton et al., 1999). A 1 mg mL^-1 gallic acid stock solution was prepared in 50% MeOH_(aq) and diluted in the same solution to yield 0–100 μg mL^-1 gallic acid standard solutions. Extracts were prepared at 1 or 2 mg mL^-1 in 50% MeOH_(aq) and serially diluted until their absorbance was within the range of the gallic acid standard curve. In a 96-well plate, 30 μL of gallic acid standard solution or extract was added to triplicate wells. To each well 200 μL of dH2O was added, then 15 μL of Folin–Ciocalteu regent. After at least 1 min, but no more than 8 min, 50 μL of 20% Na₂CO₃ (w/v) was added to all wells. The plate was mixed for 30 s on an orbital shaker, incubated at 40°C for 30 min, manually mixed with a multichannel pipette, then an additional 30 s with an orbital shaker, and finally the absorbance at 760 nm was recorded using a BioTek Cytation 3 multimode plate reader. The linear range for the assay was determined as 0–100 μg mL^-1 gallic acid equivalents (GAE), R² = 0.986. The TPC of the extracts is expressed as mg GAE/g dry extract.

Initial screening of extracts at 256 μg mL^-1 demonstrated an IC₅₀ (growth inhibition of 50% or greater) of K. fedtschenkoi extracts (1421, 1465, and 1469) against three of the ESKAPE pathogens: S. aureus, A. baumannii, and P. aeruginosa. Further testing by serial dilution assays revealed that K. fedtschenkoi extracts had IC₅₀ values ranging from 128 to 256 μg mL^-1 for these pathogens (Table 6).

Growth inhibition by dose response is reported in Figure 1. The only extract to exhibit >35% inhibition in E. faecium (EU-44) was 1468. No extracts inhibited growth of E. cloacae (CDC-08) by 20% or more. Extract 1465 (K. fedtschenkoi woody stems) exhibited >60% inhibition in A. baumannii (CDC-33) and an MIC of 256 μg mL^-1 (growth inhibition ≥ 90%) was observed against P. aeruginosa (AH-71). Extract 1508 was the only extract to exhibit at least 40% inhibition in growth at 256 μg mL^-1 against K. pneumoniae.

Human skin keratinocytes (HaCaTs) were exposed to each extract to examine possible cytotoxic effects in mammalian cells. The highest levels of cytotoxicity were observed at the 512 μg mL^-1 concentration, and ranged from 11 to 26% growth inhibition of human cells. All extracts at the 256 μg mL^-1 concentration exhibited cytotoxicity of 12% or less (Figure 2). No IC₅₀ was observed for any of the tested concentrations.

In this study, the Kalanchoe spp. extracts were screened by HPLC for the presence of several commonly occurring flavonoids; kaempferol (1) and quercetin (2), and phenolic compounds, caffeic acid (3), p-coumaric acid (4), ferulic acid (5). Both K. mortagei (extracts 1421 and 1469) and K. fedtschenkoi (extract 1468) contained 2. The extracts 1421 and 1465 of K. fedtschenkoi contained 1. The presence of 3 was also established in K. fedtschenkoi (extracts 1421 and 1465) and in K. mortagei (extract 1468) at very low levels (Figure 3).

Previous studies have shown a diverse chemistry in the genus Kalanchoe. Previous studies identified 5 in several Kalanchoe species (Gaind and Gupta, 1971; Muzitano et al., 2006a,b; Cruz et al., 2012). 1 was established in K. pinnata (Gaind and Gupta, 1971; Muzitano et al., 2006a) and K. daigremontiana (Ürményi et al., 2016). Syringic acid, 3, and 4 were identified in K. pinnata (Gaind and Gupta, 1973). A 1995 study found lupeol, lupeol acetate, β-sitosterol, and other related compounds in K. mortagei (Maiti et al., 1995).

The TFC for the Kalanchoe spp. extracts ranged from a minimum of 331 ± 33 mg GAE/g extract for 1420 to 1340 ± 116 mg GAE/g extract for 1509aq. The K. mortagei inflorescences extracts (1508 and 1509aq) had higher TFC than the other plant parts of both species, 818 mg GAE/g extract and 1340 mg GAE/g extract, respectively. The average TFC of the K. mortagei extracts with leaf and stem tissues (1420 and 1468) and the K. fedtschenkoi leaf and stem tissue extracts (1420 and 1465) were both 451 mg GAE/g extract, indicating that both species have similar TFC. However, the K. fedtschenkoi leaf and stem tissue extracts (1421, 1465, and 1469) have higher antimicrobial activities against multiple bacterial strains than the K. mortagei extracts. This suggests that the bioactivity of K. fedtschenkoi is not due to phenolic compounds.