Isolation and Antimicrobial Activities of Phytochemicals from Parinari curatellifolia (Chrysobalanaceae)

The widespread use of antimicrobial agents to treat infectious diseases has led to the emergence of antibiotic resistant pathogens. Plants have played a central role in combating many ailments in humans, and Parinari curatellifolia has been used for medicinal purposes. Seven extracts from P. curatellifolia leaves were prepared using serial exhaustive extraction of nonpolar to polar solvents. The microbroth dilution method was used to evaluate antimicrobial bioactivities of extracts. Five of the extracts were significantly active against at least one test microbe. Mycobacterium smegmatis was the most susceptible to most extracts. The methanol and ethanol extracts were the most active against M. smegmatis with an MIC of 25 µg/mL. The hexane extract was the most active against Candida krusei with an MIC of 25 µg/mL. None of the extracts significantly inhibited growth of Klebsiella pneumoniae and Staphylococcus aureus. Active extracts were selected for fractionation and isolation of pure compounds using gradient elution column chromatography. TLC analyses was carried out for pooling fractions of similar profiles. A total of 43 pools were obtained from 428 fractions. Pools 7 and 10 were selected for further isolation of single compounds. Four compounds, Pc4963r, Pc4962w, Pc6978p, and Pc6978o, were isolated. Evaluation of antimicrobial activities of Pc4963r, Pc4962w, and Pc6978p showed that the compounds were most active against C. krusei with MFC values ranging from 50 to 100 µg/mL. Only Pc6978p was shown to be pure. Using spectroscopic analyses, the structure of Pc6978p was determined to be β-sitosterol. The antifungal effects of β-sitosterol were evaluated against C. krusei in vitro and on fabrics. Results showed that β-sitosterol reduced the growth of C. krusei attached to Mendy fabric by 83%. Therefore, P. curatellifolia can be a source of lead compounds for prospective development of novel antimicrobial agents. Further work needs to be done to improve the antifungal activity of the isolated compound using quantitative structure-activity relationships.


Introduction
Infectious diseases remain the major cause of morbidity and mortality especially in the developing countries [1]. Resistance to common antimicrobial agents has led to high frequency of treatment failures and increase in treatment cost [2]. ese incidences have resulted in prolonged hospital stay, leading to an increase of nosocomial infections [3,4].
e spread of antimicrobial resistance shows that even newly designed antimicrobial agents will have brief life expectancy of efficiency [5,6].
Microbial resistance is reported to be acquired from different environmental sources including waste water, hospital surfaces, and contaminated textiles [7][8][9]. Clothes are often contaminated with certain pathogens when being worn or even after being washed [10]. Some bacteria tightly stick to the textile and tend to form biofilm which is difficult to remove [11], where resistance can be developed [7]. Contaminated textiles can cause health problems in the hospital environment [11]. e role of contaminated textiles, environmental surfaces, and medical devices in the transmission of health-associated pathogens and the development of microbial resistance has been well reported [12,13]. ere is a need for efficient therapy for bacterial diseases [14] and to search for alternative compounds that can be used as surface and textile decontaminates [15].
Medicinal plants have been identified as important sources of natural antimicrobial compounds [16] and biocides [15]. erapeutic value of these plants lies in some bioactive substances that produce a distinct physiological activity on the human body [17]. e most significant of these active phytochemicals are alkaloids, tannin, flavonoid, sterols, and phenolic compounds [18]. ese compounds can be an alternative in treating microbial infections [6], as well as in formulation of alternative disinfectants [19].
P. curatellifolia is one of the most important medicinal plants for the people of Zimbabwe [20].
is plant has various documented uses in ethnomedicine including treatment of wound infections, cancer, pneumonia, fever, bacterial infections, and inflammation [21]. A number of studies have shown that extracts from P. curatellifolia have a number of pharmacological properties ranging from antibacterial activity to anticancer activity [20]. e pharmacological properties of P. curatellifolia are linked to the presence of pharmacologically active phytochemicals including saponins, flavonoid, tannins steroid, alkaloid, cardiac glycosides, carbohydrate, balsams, and terpenes [22]. Although there are a number of studies on the pharmacological properties of P. curatellifolia extracts, there has been a paucity of documented data on antimicrobial properties of pure compounds isolated from P. curatellifolia.
To date, several methods have been utilized when preparing samples to determine the antimicrobial action of plants [1]. ese include preparation of crude extract using a variety of solvents, purification of active compounds from extracts, subsequent extraction with increasing solvent polarity, and distillation [16]. However, recent studies have shown enhanced activity of plant samples following fractionation of the extract and further purification of compounds [1,22]. e effects of β-sitosterol contaminated textiles were also assessed in an activity-guided procedure. β-Sitosterol is a natural phytosterol found in the cells and membranes of all oil producing plants [23]. β-Sitosterol has been shown to be a safe, nontoxic, effective nutritional supplement and has potential health benefits in many diverse clinical applications [24]. ese include anti-inflammatory activity, chemopreventive effects, angiogenic effects, and antidiabetic effects [25]. e aim of this study was to isolate active compounds from P. curatellifolia and determine the antimicrobial activities of the isolated compounds on some common microbes implicated in nosocomial infection [26,27] as well as on contaminated textiles. Isolation of these compounds may give a chance to evaluate the antimicrobial effects of the individual compounds as well as their synergistic effects with commercial antibiotics that are already in use. Isolated compounds from this study can then be used as templates for the production of new microbial agents or as supplements to the existing antimicrobial agents.

Plant Material.
Fresh leaves of P. curatellifolia were collected between June and July of 2017 from Norton (latitude -17.88; longitude 30.70; situated at elevation 1364 meters above sea level) in Mashonaland west Province of Zimbabwe.
e plant samples were authenticated at the National Herbarium (Harare, Zimbabwe) where a voucher specimen was made and deposited under the reference number C6E7. Permission to use the plant samples was granted by the Faculty of Higher Degrees Committee, Harare, Zimbabwe (HD/71/16).

Growth of Microbial Cells.
Four microbial isolates drawn from the list of common and emerging nosocomial pathogens were used in the study [25,26]. M. smegmatis 155 mc 2 laboratory strain was obtained from the Department of Clinical Laboratory Sciences, University of Cape Town. A clinical strain of C. krusei and K. pneumoniae (ATCC 700603) was obtained from the Department of Medical Microbiology, University of Zimbabwe. Staphylococcus aureus (ATCC 9144) was obtained from the Department of Biological Science at the University of Botswana (Gaborone, Botswana).

Culture Media and Chemicals.
All chemicals, reagents, and culture media used were purchased from Sigma-Aldrich Chemical Company (Taufkirchen, Germany). Organic solvents were distilled prior to use. K. pneumoniae and S. aureus were grown and maintained in Luria broth and Luria agar. M. smegmatis cells were grown and maintained in Middlebrook 7H9 broth supplemented with 1 g/L casein acid hydrolysate and Middlebrook 7H10 agar supplemented with 1 g/L casein acid hydrolysate. C. krusei cells were grown and maintained in Sabouraud dextrose broth and agar.

Preparation of Plant Extracts.
e leaves were washed under running distilled water to remove any dust and grit and then air-dried at room temperature. Leaf powder was prepared manually by hand pounding dry leaves using a wooden pestle and mortar. Phytochemicals from the leaf material were extracted by maceration using the DCM-MeOH extraction and the serial exhaustive extraction methods successively [28].

Extraction of Phytochemicals with Dichloromethane/
Methanol. Phytochemicals from 1500 g of powdered P. curatellifolia leaf material was extracted in 4500 mL of a solution of 50%:50% dichloromethane (DCM) and methanol, respectively, for the total extraction. e mixture was left for extraction for 72 hours with constant shaking. e extraction mixture was filtered through cotton wool and then through Whatman No. 1 paper. e leaf debris (residue) was air-dried and stored at room temperature for further extraction. e filtrate was concentrated using a rotary evaporator RII (BUCHI, Labortechnik AG, Switzerland) at 50°C and low pressure. e extracts were dried to a constant mass under a fan in a fume hood cabinet. Dried extracts were stored in sterile screw cap tubes at 4°C until use.

Serial Exhaustive Extraction.
Extraction was done sequentially using different solvents on the leaf debris (residue) from the dichloromethane-methanol mixture (50% v/v) (DCM-MeOH). e serial exhaustive extraction was done in the following sequence of solvents: hexane, DCM, acetone, ethyl acetate, ethanol, methanol, and water. At each extraction stage, the leaf debris generated after filtration was left to dry in a sterile fume hood; after drying, the debris was used in the subsequent extraction stage. e filtrate was concentrated using a rotary evaporator RII (BUCHI, Labortechnik AG, Switzerland) at 50°C and low pressure. e extracts were dried to a constant mass under a fan in a fume hood cabinet. Dried extracts were stored in sterile screw cap tubes at 4°C until use. e weight of dry extract obtained was recorded and percentage yield was determined. extraction yield(%) � weight of dried plant extract (g) weight of dried plant material(g) × 100%. (1) Crude extracts were collected in vials for further analysis.

Preparation and Standardization of Inocula.
e inoculum of each microbe was prepared by reviving cell from glycerol stocks. Cells for the assay were standardized according to the 0.5 McFarland solution, which is equivalent to 1.5 × 10 8 CFU/mL.

INT Colorimetric Assay for Minimum Inhibitory Concentrations (MIC).
e inhibitory potential of microbial growth by both the crude extracts and the pure compound was determined by broth microdilution method in 96-well microtiter plates [29]. e MIC assay was conducted using 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyl-2H-tetrazolium chloride (INT) colorimetric assay according to Eloff [30] with slight modifications. e assay for minimum inhibitory concentration was conducted according to the plate layout shown in Figure 1. Working concentration of test samples was prepared by dissolving 0.004 g of the extracts of isolated compound in 1 ml dimethylsulfoxide (DMSO) to make a concentration of 4 mg/ml. en, 0.5 ml was taken from the working concentration and made up to 10 ml by adding sterile broth media to make a test concentration of 200 µg/ ml. DMSO was used to prepare the samples was 2.5%. A serial twofold dilution was performed to obtain extract concentrations ranging from 0.4 to 200 µg/mL using sterile broth. Liquid cultures of each test microbe were grown in broth media. ese were diluted in fresh broth, and 100 µL was applied to the wells of a 96-well plate. In each case, approximately 1.5 × 10 8 CFU/mL of exponentially growing cells were inoculated for each strain. e extracts (100 µL) were added to these wells in decreasing concentrations and mixed by pipetting. Each well was further diluted with 100 µL microbial inoculum. e final extract concentration in the 96-well plate ranged from 0.2 to 100 µg/mL. Cells in wells with unexposed broth were used as the positive control for microbial growth, while wells with plain broth were used as negative cell growth ( Figure 1). Cell density of the plate was measured at 590 nm using a microplate reader (Tecan GENios Pro, Grödig, Austria) prior to incubation. e plate was incubated at 37°C for 18 hrs. Cell density was measured after incubation. Growth of cells was determined by calculating the difference of the preincubation value from the postincubation value. Data are presented as percentage inhibition of inoculum. Percentage inhibition was obtained using the following equation: percentage inhibition(%) � positive control value − sample value positive control value × 100.
Following incubation, 50 µL of 0.2 mg/mL INT was added in each well. e plates were incubated for an additional 30 min at 37°C. Viable microbes reduced the yellow dye of INT to a pink color. MIC values were defined as the sample concentration that exhibited complete inhibition of bacterial growth, that is, the sample concentration that prevented the color change of INT [30].

Minimum Bactericidal/Fungicidal Concentration (MBC/MFC) Determination.
e MBC/MFC was determined after MIC assays [31]. e MBC/MFC of the sample was determined by subculturing 50 µL of the suspensions from the wells that did not show any growth after incubation during MIC assays on an agar plate. e inoculated agar plates were incubated at 37°C for 24 hours before observation. e MBC/MFC was defined as the lowest concentration of a sample that completely inhibited the visible microbial growth on the agar plates. Each assay was performed in triplicate.

Chromatographic Separation of Extracts.
Active extracts against at least one of the test organisms were selected and combined. Combined extracts were fractionated using the gradient elution column chromatography. In this study, the methanol extract, DCM:MeOH extract, ethanol extract, and hexane extract were selected for chromatographic separation.

Gradient Elution Column Chromatography.
Selected extracts were combined and adsorbed on equal mass of silica (70 g of silica, 60-120 mesh) to make a loading Advances in Pharmacological and Pharmaceutical Sciences sample. e sample was packed on a silica gel column, and the gradient elution method was used to run the column. e column was first eluted with n-hexane as the mobile phase, and the polarity of the mobile phase was increased by gradually adding 2.5% of ethyl acetate increments. After reaching 100% ethyl acetate, the polarity was further increased by gradually adding 2.5% methanol. Volumes of 250 mL fractions of the eluate were collected during the process. A total of 428 fractions were obtained from the column chromatography, and each fraction was concentrated using a rotary evaporator RII (BUCHI, Labortechnik AG, Switzerland). e fractions were dried to a constant mass under a fan in a fume hood cabinet. All fractions were stored in sterile screw top tubes at room temperature.

TLC Profiling of Chromatographic Fractions.
Concentrated fractions were subjected to thin layer chromatography (on aluminium plate coated with Mercy, silica gel 60 F254). TLC plates were developed for each fraction in a saturated tank with a mobile phase of slightly higher polarity than used in column chromatography to elute the fraction. For example, fractions eluted in 100% hexane were developed in 97.5:2.5 (HEX:EA) solution. Developed chromatograms were located by observation under short and long ultraviolet light (short UV and long UV at 254 nm and 360 nm, respectively), then sprayed with 10% w/v sulphuric acid, and dried at 100°C. Fractions showing similar profile were pooled into the same pool. Pooled fractions were dried in glass vials and their yields determined. Pooled fractions were then kept at room temperature for further analysis.

Purification of Compounds Using Preparative TLC.
Fractions showing the best TLC separation as well as sufficient yield were selected for further purification of single compounds using the preparative TLC. In this case, pools 7 and 10 were selected for the isolation of single compounds. Preparative TLC was made by absorbing the extracts on 2 mm layer of silica Type G with 13% calcium sulphate on 20 × 20 cm glass plate. e plate was activated by heating in oven at 60°C for 30 minutes prior to use. e selected pools were combined, dissolved in chloroform, and then loaded on the activated TLC plate. e plate was allowed to develop in a chloroform-saturated tank. Once developed, the separated chromatograms were observed under UV light at 254 nm using a UV lamp in a UV-viewing cabinet. At least four chromatograms of different colors were separated. Silica at the position showing a single isolated chromatogram on the plate was scraped off using a metal spatula blade. e chromatogram content on the scraped silica was eluted using chloroform through a Whatman No. 1 filter paper into four different glass vials. e eluent was then evaporated at room temperature and then stored as solids in the vials. e crystallized solids in the vials were further analyzed for purity using TLC with chloroform as the mobile phase. e solids that gave a single TLC chromatogram confirmed the purity of the isolated compound. Using TLC, only one compound out of four was confirmed to be a pure compound. Antimicrobial evaluation of the pure compound was determined against the test microbes. NMR analyses and determination of the mass of the isolated compound were carried out.  . e "cells" column is the positive control for microbial growth (cells were inoculated in untreated broth media). e "media" column is the negative control for cell growth (wells contained uninoculated broth media). e "antibiotic + cells" rows are the antibiotic/antifungal positive control (cells were inoculated in wells with decreasing standard antimicrobial/antifungal concentration). e "extract + cells" rows are the test wells (cells were inoculated in wells with decreasing extracts concentrations).

NMR Analyses and Determination of the
shifts for 1 H NMR and 13 49.0 ppm). e 2D NMR experiments were run using standard pulse sequences. Molecular formulae were determined by electrospray ionization with a 7-T hybrid ion trap and a TOF detector running in positive or negative mode. β-Sitosterol ( Figure 2) was identified as one of the pure compounds, and its effects on the test organisms were evaluated.

Biocidal Activity Assays of the Most Active Compounds on
Mendy Fabric. Biocidal evaluation of the purified compounds was done on the Mendy fabric, which is the material used to make uniforms for nurses in Zimbabwe. e biocidal evaluation was done using a simple and rapid method for optical visualization and quantification of bacteria on fabrics using the INT dye, by Stiefel et al. [10] with some modifications. Absorbance measurements for the biocidal activity assay were done using a Shimadzu UV/VIS UV-1601 spectrophotometer (Shimadzu, Kyoto, Japan). Standard curves to correlate the staining intensity with the numbers of microbial cells were first generated. e standard curves were used to determine the number of bacterial and fungal cells left after the microbial contaminated fabric was treated with the isolated compound.
To generate the standard curves, 100 µL microbial culture of different concentrations (OD 600 : 0, 0.125, 0.25, 0.375, and 0.5) was directly applied to the fabric samples in 6-well plates. Samples were incubated for an hour at 37°C. After incubation, 1 mL of media and 0.2 mg/ml 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyl-2H-tetrazolium chloride (INT) was added to each fabric (staining with tetrazolium salt). e sample was incubated for 30 minutes at 37°C and with shaking at 40 rpm. After incubation, the liquid was removed, and 1 mL dimethyl sulfoxide (DMSO) was added to each well. e dye was removed by pipetting the mixture several times onto the fabric until a homogenous color was obtained. Absorption of the eluate was measured at 490 nm using DMSO as a blank. As a background control the different fabrics were treated in the same way using media without bacteria.
To correlate OD 600 values of the applied culture with cell numbers, colony forming units (CFU) were determined at the same time point as the staining of bacterial culture. e same bacterial solutions as added to fabrics were also incubated for 1 hour at 37°C and 80 rpm. After 1 hour, the bacterial culture was used to determine cell numbers by plating a 1:10 dilution series on agar. e CFU of appropriate dilutions were counted after 1 day, and the exact cell numbers corresponding to the applied culture were determined. e bacterial culture with an OD 600 value of 0.5 corresponded to about 3.15 × 10 10 bacteria per square centimetre of fabric for M. smegmatis cells and 5.6 × 10 8 fungal cells per square centimetre for C. krusei. All the standard curves were generated using GraphPad Prism (Version 5.03, GraphPad Software Inc., San Diego, California, USA).
To evaluate the biocidal activities of the compounds on the Mendy fabric, 100 µL of the microbial culture with an OD 600 of 0.5 in broth media was directly applied to the fabric samples in 6-well plates and incubated for 1 hour at 37°C. After incubation, 100 µL of (200 µg/mL) test compound was directly applied to the fabric samples in the 6-well plates and incubated for 30 minutes at 37°C. Concentrations of 100 µg/ mL and 50 µg/mL of chlorhexidine digluconate were applied to pieces of the same fabric in different wells as the biocide positive control. After incubation, 1 mL growth medium containing 0.2 mg/mL INT was added to the fabric without removing unbound bacteria and incubated for 30 minutes at 37°C and 40 rpm. e dye was eluted from the fabric using 1 mL DMSO and pooled with the supernatant, in order not to lose the unbound bacteria. Staining intensity was quantified by measuring absorbance at 490 nm using a Stat Fax 2100 microplate reader (Awareness Technologies Inc., Westport, USA). For background control, media without bacteria were used during incubation. e number of viable cells that remained on the fabric was quantified by comparing the measured signal intensity to the standard curve.

Statistical
Analysis. Statistically significant differences between the mean of the controls and the tests were analyzed using one-way ANOVA with Dunnett's multiple comparison posttest. Enzyme kinetics were analyzed with linear regression using GraphPad Prism (Version 5.03, GraphPad Software Inc., San Diego, California, USA).

Extraction Yields of the Plant Extracts.
e highest yield, 3.53%, was obtained from extraction of P. curatellifolia leaves with the dichloromethane (DCM) and methanol, while the least yield, 0.23%, was from the hexane extraction (Table 1).

In Vitro Antimicrobial Activity of the Extracts.
Evaluation of the antimicrobial activities of the extracts against the four test organisms was done using the INT colorimetric assay for minimum inhibitory concentrations. e study has shown that all the extracts used in the study exhibited varying degrees of antimicrobial activity against all the microorganisms tested (Table 2). e DCM-methanol extract had the highest activity against S. aureus, reducing cell growth by 90% after incubation. However, the ethyl acetate extract had the least activity of only 3.6% against S. aureus (Figure 3(a)). ere was no complete inhibition of growth for S. aureus and K. pneumoniae by any of the crude extracts even at the highest concentration, 100 µg/mL, that was tested (Figures 3(a) and 3(b)). However, the DCMmethanol extract had the highest activity, 75%, against the K. pneumoniae, while the water extract had the lowest activity, only 26.9%, after incubation (Figure 3(b)). M. smegmatis was more susceptible to all the P. curatellifolia extracts compared to other test microbes (Table 2). e ethanol and methanol extracts had the highest activity of MIC, 25 µg/mL, against the M. smegmatis, followed by the DCM-methanol and the acetone extract which had MIC of Advances in Pharmacological and Pharmaceutical Sciences 5 50 µg/mL each on the same cells (Figure 3(c)). e DCM-MeOH had an MBC value of 50 µg/mL while the methanol extract had an MBC value of 100 µg/mL against M. smegmatis (Table 2). Only the hexane extract and water extract were able to inhibit the growth of C. krusei cells by more than 50% at a concentration of 100 µg/mL after incubation ( Figure 3(d)). e DCM-methanol and the acetone extract completely failed to inhibit growth of C. krusei (Figure 3(d)). e hexane extract was the most effective in inhibiting growth of C. krusei with an MIC value of 25 µg/ mL after incubation (Figure 3(d)), but no extracts exhibited MFC against C. krusei (Table 2).

Chromatographic Separation of Extracts.
e extracts that had potent antimicrobial activity are as follows: DCMmethanol extract, hexane extract, ethanol extract, and ethanol extract by inhibiting growth of at least one test organism (Table 2) were selected for chromatographic separation. Selected extracts were combined and fractionated using gradient elution column chromatography. A total of 43 distinct pools were obtained from 428 fractions. e pools were numbered from 1 to 43. Pools 7 and 10 ( Figure 4) were observed to exhibit better separation of chromatogram on TLC plate. Pools 7 and 10 were selected for further separation of single compounds using preparative TLC. At least four chromatograms with different colors, red, white, dark red, and purple, were obtained. e chromatograms were assigned code names Pc4963r, Pc4962w, Pc6978o, and Pc6978p, respectively, for identification ( Figure 5). e migrating chromatograms were isolated by scrapping off silica containing the targeted chromatograms from the glass plate. e composition of separated chromatograms on the scraped silica was eluted using chloroform through the Whatman No. 1 filter paper into different glass vials.
e purity of each isolated chromatogram was determined using TLC. At least four compounds were obtained after elution of the four chromatograms from the silica. e four compounds showed different colors, red, white, orange, and purple, with Rf values of 0.74, 0.40, 0.57, and 0.38, respectively ( Figure 6). e compounds were given code names Pc4963r, Pc4962w, Pc6978o, and Pc6978p ( Figure 6). Only the contents of Pc6978p were shown to contain a single compound by showing a single spot on the TLC plate ( Figure 6). e structure of the pure isolated Pc6978p was determined to be β-sitosterol ( Figure 2) by NMR as confirmed by the 1 H NMR (CDC13, 500 MHz) and 13 C NMR (CDCl 3 , 125 MHz) spectrum of Pc6978p (Figure 7). e compounds showing more than one spot on the TLC were stored in glass vials for further analysis on a later stage.

In Vitro Bioactivities of the Purified Compounds.
Compounds Pc4963r, Pc4962w, and Pc6978p were selected for the evaluation of antimicrobial activities against the test organisms. Compound Pc6978o was left out due to very low yield, lower than 3 mg (Table 2). Figure 8 shows a graph of the antimicrobial activities of Pc4963r, Pc4962w, and Pc6978p at concentration of 100 µg/mL against test cells. e cell densities correspond to the growth of cells after incubation at 37°C overnight. Compound Pc6978p showed the highest reduction effects on the growth of S. aureus cells with a percentage antibacterial activity of 75.3%, followed by Pc4963r which had a percentage activity of 71.1% (Figure 8(a)). Compared to the other two isolated compounds, Pc4962w had the least antibacterial activity against S. aureus, with a percentage activity of 50.5% (Figure 8(a)). However, statistical analysis using the one-way ANOVA (where P < 0.05) has shown that all compounds had statistically significant reduction effects on growth of S. aureus (Figure 8(a)). Data analysis showed that all the compounds significantly reduced the growth of K. pneumoniae (Figure 8(b)). However, compared to the other two isolated compounds, Pc6978p had the least antibacterial activity against K. pneumoniae, with a percentage activity of 39% (Figure 8(b)). Compound Pc4962w showed the highest reduction effects on the growth of K. pneumoniae cells with a   Advances in Pharmacological and Pharmaceutical Sciences percentage antibacterial activity of 51%, followed by Pc4963r which had a percentage activity of 44% (Figure 8(b)). Compound Pc6978p showed the highest reduction effects on the growth of M. smegmatis cells with a percentage antibacterial activity of 97.9%, followed by Pc4963r which had a percentage activity of 46.8% (Figure 8(c)). Compared to the other two isolated compounds, Pc4962w had the least antibacterial activity against M. smegmatis with the percentage activity of 31.9% (Figure 8(c)). However, statistical analysis using the one-way ANOVA (where P < 0.05) has shown that all the compounds significantly reduced the growth of the M. smegmatis (Figure 8(c)). e antifungal activities of Pc4963r, Pc4962w, and Pc6978p at concentration of 100 µg/ mL against C. krusei cells were investigated. e graph in Figure 8(d) shows that all the three isolated compounds completely inhibit the growth of C. krusei cells.

Biocidal Activity Assays of β-Sitosterol (Pc6978p) on
Mendy Fabric. Pc6978p had the highest activity against C. krusei and M. smegmatis. erefore, biocidal activity of compound Pc6978p was evaluated against M. smegmatis and C. krusei. Standard curves were first generated by adding different concentrations of cell on different pieces of fabric; the increase in color intensity corresponded to the increasing number of cells (Figure 9). A typical layout of the fabric used to generate the standard curves is shown in Figure 9(a). e color intensity of the dye on fabric was increasing from OD 490 0 to OD 490 0.5 indicating the increasing cell numbers attached to different pieces of Mendy fabric. Figures 9(b)   Mendy fabric (Figure 10(a)). e compound removed and killed about 4.45 × 10 8 C. krusei cells per cm 2 of Mendy fabric after 30 minutes of incubation. e positive control chlorhexidine digluconate 50 µg/mL and 100 µg/mL had 59% and 79% biocidal activity on C. krusei attached to the Mendy fabric ( Figure 10). Statistical analysis using the one-way ANOVA showed that there was no difference between the biocidal activities of the positive control chlorhexidine digluconate at 100 µg/mL and the compound Pc6978p. Pc6978p did not have significant biocidal activity, only producing 5.8% cell reduction against M. smegmatis attached to the Mendy fabric ( Figure 10(b)).
However, the compound has been shown have biocidal activity of around 1.9 × 10 9 M. smegmatis cells attached to 1 cm 2 of Mendy textile after 30 minutes of incubation. e positive control chlorhexidine 50 µg/mL and 100 µg/mL had 64.7% and 90.1% biocidal activity on K. pneumoniae cells attached to the Mendy textile in only 30 minutes, which was significantly different from the activity of the compound Pc6978p (Figure 10(b)).

Structure Determination of the Isolated Compound.
e structure of the compound Pc6978p was confirmed by the nuclear magnetic resonance (NMR) spectroscopy to be β-sitosterol ( Figure 2) [31,32]. e following are the spectroscopic characteristics of the compound Pc6978p: 1 (Figure 7(a)).

Discussion
In this study, β-sitosterol was isolated from P. curatellifolia extracts during extraction to determine active antimicrobial and biocidal compounds from this plant. Both fungi and bacteria remain major life threatening pathogens in humans, mostly to the immunocompromised or immunodeficient patients [33]. Despite the existence of potent antibiotic and antifungal agents, resistant and multiresistant strains are continuously emerging, imposing the need for a permanent search for alternative microbial agents [34]. Moreover, disinfection proved to be a frontline strategy of controlling microbial contamination and spread and development of resistance [35]. erefore, this study was made to assess both the microbial and biocidal activity of phytochemicals from P. curatellifolia, as plants extracts are considered to be one of the best sources of alternative antimicrobials [36].
However, it is known that the activity of the plant extracts could be due to the different classes of phytochemicals present, their quantities, and the possible interactions with other constituents of the extract [1]. Furthermore, antimicrobial properties of plant extracts were observed to be influenced by several factors including the sample preparation and the extraction process [16,36]. erefore, in the study, precautions were taken to ensure that potential active phytochemicals were not lost, distorted, or destroyed during the extraction. ese include air-drying the sample at room temperature rather than oven drying at 60°C [16]. e effects of heat on potential active compounds were demonstrated by the significant reduction in the active anthocyanin content in some leaf extracts, possibly due to accelerated chalcone formation on exposure to high temperatures [37]. It was reported that bioactive phytochemicals, such as sinensetin and rosmarinic acid content, were affected by the oven and sunlight drying [38]. * * * * * * * * * * * * M. smegmatis (c), and C. krusei (d). "Media" column is the negative control showing no cell growth. "Cells" column is the positive control for cell growth. Error bars represent the standard deviation from mean (n � 4). e asterisk ( * ) indicates the significant difference from the positive control using statistical one-way ANOVA (where * P < 0.05, * * P < 0.01, * * * P < 0.001, * * * * P < 0.0001).
Maceration extraction technique based on the extracting power of different solvents was utilized in this study [16].
is technique targeted extraction of a wide range of compounds from the leaf material; hence, a combination of the total extraction (DCM-MeOH) and the serial exhaustive method was used on the same sample of P. curatellifolia. e DCM-MeOH extraction had the highest percentage yield, 5.53%. is extraction technique uses a cosolvent extraction system of dichloromethane (nonpolar) and methanol (polar). e method was designed to extract both the polar and nonpolar compounds in a single extraction process, which justify the high yield compared to the polarity specific extraction methods [39].
In the polarity specific serial exhaustive methods used in this study, the methanol extraction had the highest yield, 3.31%, followed by the water extract that had 3.0%. e general observation was that the extraction yield increased as the polarity of the solvents increased. Increase in yield may be an indication of the presence of larger amounts of polar compounds in P. curatellifolia leaves compared to nonpolar compounds. Higher amounts of polar compounds are in accordance with previous observations that the dominant compounds in plants leaves are polar in nature [40].
Bhunu et al. [28] used a similar extraction method to extract phytochemicals from P. curatellifolia. Similar to the result shown in this study, their study [28] has also shown that the DCM-MeOH extraction had a higher yield compared to the serial exhaustive extraction. Reduction in yield of serial exhaustive extraction in this study may be probably due to the fact that the serial exhaustive extraction was done on the same plant materials that have some or most of the extracts already extracted by the DCM-MeOH extraction.
Furthermore, it was noted that the yield of DCM-MeOH extraction and the combined yield of the serial exhaustive extraction in the study by Bhunu et al. [28] were higher than the yields produced from the same extraction in the current study. eir study [28] has shown that the DCM-MeOH extraction yield was as high as 18% compared to 3.5% percent of this recent work. ey also produced 12% of the combined serial exhaustive extraction compared to only 9% from the current study. Reduced yield in the current study is probably due to the material to solvent ratio of the extract. Bhunu et al. [28] used 1:10 (w/v) ratio for extraction and produced better extraction yield compared to 1:3 (w/v) ratio used in this study.
ese observations have justified the importance of careful selection of the extraction methods and solvent ratios for better phytochemicals extraction [41]. Since these two studies have used almost the same extraction methods and solvents produced different yields. e observation was that extraction efficiency in maceration of extraction is also affected by the solubility of the solid plant material and effective diffusion in the liquid extraction solvent [42]. e higher the solid to solvent ratio, the higher the yield [42]. Moreover, all stages of extractions, from the pre-extraction and extraction, are equally important in studying plant phytochemicals [41]. Sample preparation, such as grinding and drying, was also shown to affect the efficiency and phytochemical constituents of the final extractions, which eventually have an effect on the final extracts [16].
In vitro bioactivities of all the seven crude extracts were evaluated against the Gram-positive S. aureus, the Gramnegative K. pneumoniae, the acid-fast M. smegmatis, and the fungus C. krusei before fractionation and isolation of single compounds. e choice of purification of compound used in this study was activity-guided. According to Kuete et al. [43], antibacterial activity of a crude plant extract has been defined as significant when MIC is below 100 µg/mL, moderate when 100 µg/mL < MIC > 625 µg/mL, or low when MIC > 625 µg/mL. e antimicrobial activity of a compounds is defined as significant when MIC is below 10 µg/ mL, moderate when 10 µg/mL < MIC > 100 µg/mL, or low when MIC >100 µg/mL [43]. erefore, the bioactivity assay in this study was guided by those standards. Extracts were considered to be significantly active when MIC values below 100 µg/mL were obtained on at least one tested organism. Interestingly, 5 out of 7 extracts were active against at least one test microbe.
M. smegmatis was the most susceptible to most of the P. curatellifolia leaf extracts, with the ethanol and methanol extract being the most active extracts against M. smegmatis cell growth, with an MIC of 25 µg/mL. e DCM-MeOH extract and the acetone extract had an MIC of 50 µg/mL on M. smegmatis. e activities of extracts against M. smegmatis were observed to generally increase as extraction solvent polarity increased from hexane to methanol. Out of the four extracts that inhibited the growth of M. smegmatis cells, only the DCM-MeOH and the methanol extract had MBC values of less than or equal to 100 µg/mL against M. smegmatis. is observation indicates possible bactericidal effects of both the DCM-MeOH and the methanol extract again M. smegmatis at concentration of 100 µg/mL, since the two extracts completely inhibited the visible M. smegmatis growth on the agar plates after incubation at 37°C for 24. e acetone and ethanol extracts were bacteriostatic against M. smegmatis at a concentration less than or equal to 100 µg/mL. Significant bactericidal effects of DCM-MeOH extracts against M. smegmatis may be because of synergism of different compounds since the DCM-MeOH extracts contained a wide range of compounds including both polar and nonpolar compounds.
Bhunu et al. [28] also demonstrated susceptibility of M. smegmatis cells to P. curatellifolia leaf extracts. According to their study [28], the MIC of the P. curatellifolia leaf extracts was 6.2 µg/mL for the acetone extract, 12.5 µg/mL for both the ethanol and the DCM-MeOH extracts, and 50 µg/mL for both the methanol and ethyl acetate extracts.
e MIC values from the study by Bhunu et al. [28] were slightly lower than the value obtained in this study. is may be due to factors including the effectiveness of the extraction and preparative methods [16]. Susceptibility of M. smegmatis to P. curatellifolia leaf extracts may be due to the destruction of waxy cell envelop on mycobacteria [42]. e waxy cell envelop is known to be innate, making the mycobacteria the least susceptible to antimicrobial agents [42]. However, replicates. e asterisk ( * ) indicates the significant difference from the positive control chlorhexidine 100 µg/mL using statistical one-way ANOVA (where * P < 0.05, * * P < 0.01, * * * P < 0.001, * * * * P < 0.0001, and "ns" means no significant deference). some phenolics, e.g., the gallic acid, were demonstrated to interact with these membranes envelops [42]. e hexane extract was among the extracts that showed significant antimicrobial activity according to [43] standards. Despite being less potent to all the bacterial test organisms, the hexane extract showed significant antifungal activity against C. krusei. is may be an indication that nonpolar extracts of P. curatellifolia have less antibacterial activities but stronger antifungal. e MIC of the hexane extract on C. krusei was 25 µg/mL. e extract was not fungicidal at a concentration less than or equal to 100 µg/mL. Interestingly, of all the extracts, only the hexane extract showed potent antifungal activity. Silva Sa et al. [44] have shown that nonpolar fractions including the hexane and dichloromethane fraction from Myrcia tomentosa (Aubl.) DCM leaf extracts showed antifungal activity against Candida species with low MIC values, which ranged from 4 to 256 µg/mL. erefore, it is most likely that the constituents of the nonpolar extracts are responsible for the anti-Candida activity of P. curatellifolia. e probable reason for the strong activity of hexane extract against the C. krusei may be the interaction between C. krusei cells and nonpolar compounds, since the strong hydrophobicity in the C. krusei cell membrane allows stronger affinity to nonpolar compounds [45]. In a similar study, Samie et al. [46] reported that hexane extracts were among the active extracts against C. krusei. Samie et al. [46] went on to conclude that most Candida species were susceptible to plant hexane extracts suggesting that many of the antifungal components of these plants were nonpolar compounds [45]. Lack of significant antibacterial activity by the hexane extract with some solvents showing activity is in accordance with the observations by Peni et al. [22], who showed that different solvent extracts of the same plant may have different pharmacological properties.
None of the extracts used in this study were able to completely inhibit both the Gram-positive S. aureus and Gram-negative K. pneumoniae at concentration less than or equal to 100 µg/mL. Lack of activity below 100 µg/mL of extracts against S. aureus was previously shown by Oshomoh and Idu [47], who indicated that the least MIC of some P. curatellifolia extracts was 3.125 mg/mL against S. aureus. However, the results from this study have shown that the DCM-MeOH extract had considerable activity against Gram-positive cells at a concentration of 100 µg/mL. At that concentration, the DCM-MeOH extract inhibited up to 90% of the growth of S. aureus cells. e activities of most extracts against S. aureus were generally above 50%. However, the higher activity against S. aureus in this study was generally observed with polar P. curatellifolia extracts. ese observations were in agreement with those by Peni et al. [22]. e DCM-MeOH had the highest activity against the Gram-negative K. pneumoniae inhibiting up to 74.8% of cell growth. A previous study by Halilu et al. [48] showed that the MIC of P. curatellifolia extracts against some Gram-negative microbes was above 100 µg/mL limit. e least MIC of P. curatellifolia extracts against P. aeruginosa was 1.5 mg/mL [48]. According to Halilu et al. [48] P. curatellifolia extracts were more active against Gram-positive compared to the Gram-negative microbes. e results from the study by Halilu et al. [48] agree with those from this study, showing that the DCM-MeOH leaf extracts from P. curatellifolia had a higher activity against S. aureus (89.9%) compared to K. pneumoniae (74.5%). e susceptibility of the Gram-positive bacteria may be due to the disruption of the cell walls by some phytochemicals present in the crude extracts commonly classified as phytoanticipins [49].
After evaluating the activity of the crude extracts, extracts that showed considerable activity against at least one of the test microbes were selected for fractionation and isolation of pure compounds. e hexane, methanol, ethanol, and DCM-MeOH were selected for fractionation. A chromatogram with more than a single spot on the TLC was considered not pure. e purple compound Pc6978p (Rf 0.38) was considered pure since only a single spot was observed on the TLC plate. After elution of the chromatograms from the preparative TLC silica, compound Pc6978p was concentrated to a white crystalline powder. Pc4963r, Pc4962w, and Pc6978o were concentrated to green crystalline powder, white crystalline powder, and green crystalline powder, respectively. e structure of Pc6978p was determined to be β-sitosterol on the basis of 1 H and 13 C NMR spectroscopic data and by comparing them to those reported in the literature [50].
β-Sitosterol is one of most spread phytosterols in the plant kingdom [23]; β-sitosterol was isolated previously from a number of plant species [51][52][53], but to the best of our knowledge, β-sitosterol was isolated for the first time from P. curatellifolia extracts. In vitro bioactivities of β-sitosterol were evaluated against the four test organisms. e isolated β-sitosterol failed to show significant activity against all the test organisms according to Kuete [43] proposed standards. Nonetheless, the compound managed to reduce the growth of M. smegmatis cells by 97.9%. ere is a notable loss of activities of Pc6978p compared to the activity of some crude extracts (DCM:methanol, acetone, ethanol, and methanol) against M. smegmatis. is may be attributed to loss of synergism of compounds in the crude extracts through purification. Synergistic effects occur when two compounds increase one another's potency, resulting in mixtures that have stronger effect than predicted based on activities of their components in isolation [54]. High antimicrobial activity of some crude extracts observed may be due to the synergistic effects of the compounds present in the extracts [54]. For example, combinations of thymol and eugenol were shown to have synergistic effects against E. coli [55].
β-Sitosterol reduced the cell growth of the Gram-positive S.aureus and Gram-negative K. pneumoniae by 75% and 39%, respectively, at 100 µg/mL. It was also noted that the isolated β-sitosterol has significantly reduced activity against these two microbe as compared to the activity of the DCM: methanol and methanol extracts. is may be an indication that the most active compounds against Gram-positive and Gram-negative bacteria may be more in polar extracts (e.g., methanol extracts) than in nonpolar ones, since some studies have reported the isolation of β-sitosterol in nonpolar solvents [23]. Loss of activity of the isolated β-sitosterol may also be due to loss of synergistic effects of compounds present in extracts against S. aureus and K. pneumoniae.
While the activity of β-sitosterol cannot be classified as significant [43], results from this study showed significant activity of this compound against all the test bacterial strains. e observed antibacterial activity of the isolated compound are in agreement with the reported antibacterial activity of β-sitosterol against the same bacterial strains [52]. Using the disc-diffusion method [52], other researchers have also reported low activity of β-sitosterol against both the Gramnegative and Gram-positive cells. e growth inhibition in that study was 11 mm and 13. 5 mm against S. aureus and Gram-negative E. coli, respectively, using the disc-diffusion method [52]. ere are also reports that show antibacterial activity of β-sitosterol against several bacterial species [23,56]. erefore, from this study, it was observed that β-sitosterol has activity against the test bacterial cells, although the activity can only be classified as low according to Kuete [43]. Low but statistically significant activity of the β-sitosterol against Gram-positive, Gram-negative, and acid-fast bacteria has highlighted a degree of the broadspectrum activity of the β-sitosterol. However, of all the activities of β-sitosterol, the lowest observed was against Gram-negative bacteria.
β-Sitosterol had moderate activity against C. krusei with an MIC value of 25 µg/ml and MFC of 50 µg/mL against C. krusei.
e results confirm reports suggesting that phytosterols do have an antifungal potential [57]. From this study, an increase in in activity against C. krusei was observed between the activity of the most active hexane extract and activity of the purified compound. e hexane extract failed to completely inhibit the growth of C. krusei after incubation at a concentration of 100 µg/mL for 18 hrs, while the pure compound was able to kill all the cells at 50 µg/mL in the same incubation conditions. e enhancement of antimicrobial activity of β-sitosterol as a single entity chemical may be due to the elimination of the antagonistic effects of different compounds found in crude extracts. Antagonistic effects occur when two compounds inhibit one another's activities, such that mixtures are less effective than predicted based on the activities of each compound in isolation [55]. Antagonistic effects of different compounds in extracts are known to reduce the activity of some active compounds [54]. For example, eugenol and thymol had antagonistic effects against Crithidia fasciculata [58]. Hence, purification of some compounds is sometimes recommended to achieve the highest antimicrobial activity of some phytochemicals [58]. e antifungal activity of β-sitosterol observed in this study is in agreement with a previous study reporting that β-sitosterol inhibited the growth of C. albicans and C. krusei at 50 µg/mL [59]. Although some studies have indicated that β-sitosterol was fungistatic rather than fungicidal [24], this study has shown that β-sitosterol was fungicidal at 50 µg/mL. However, from these observations, further analysis of β-sitosterol on the anti-Candida species, especially C. krusei, may be necessary since this fungal species is clinically significant. C. krusei induces candidiasis which is also caused by many other species of the Candida genus [60]. C. krusei is also listed as emerging fungal nosocomial pathogen whose infection is common among immunocompromised patients and those with hematological malignancies [60]. erefore, further analysis may be useful to determine if the β-sitosterol can be used to treat such fungal related ailment, as well as to unlock more of its value in human health and hygiene. For example, its synergistic potential can be used to enhance the activity of already present antimicrobial agents to minimize microbial resistance. e isolated β-sitosterol can also be used as an additive to present antifungal agents. Additive effects indicate that the effects of chemicals are independent of one another [55] but have similar modes of action, such that adding a second compound has the same effect as adding more of the first compound [61]. A clinical example of additive effects due to independent actions of phytochemicals, artemisinin, and curcumin against malaria was descried by Nandakumar et al. [62].
Chemical composition and structures of the contents in Pc4963r, Pc4962w, and Pc6978o could not be determined in this study. However, antimicrobial activities of Pc4963r and Pc4962w were evaluated against all the test organisms. Antimicrobial activities of Pc6978o were not evaluated due to very low yield. e highest activities of both Pc4963r and Pc4962w were observed against C. krusei cells; both compounds showed moderate activity against C. krusei. Pc4962w had MIC and MFC values of 25 µg/mL and 100 µg/mL, respectively. Pc4963r had MIC and MFC values of 50 µg/mL and 100 µg/mL against C. krusei. e study has shown an increase in activity against C. krusei between the activity of the most active hexane extract and the activity of both Pc4963r and Pc4962w. e hexane extract failed to totally eliminate the C. krusei cells following incubation at a concentration of 100 µg/mL for 18 hours. On the other hand, Pc4963r and Pc4962w were fungicidal cells at 100 µg/mL under the same incubation conditions. Improvement in activity of both compounds may be due the elimination of the antagonistic effects of different compounds found in crude extracts. e similarity of these results to those of β-sitosterol coupled with the fact that β-sitosterol, Pc4963r, and Pc4962w were isolated from closely related pools may lead to the suggestion that both Pc4963r and Pc4962w may also be phytosterols. is suggestion has been derived from a previous suggestion that most phytosterols have an antifungal potential [49].
After determination of the biological activity of the isolated compound, the study went on to determine the applicability of the obtained results. erefore, following the activity-guided approach of analysis, biocidal effects of Pc6978p were evaluated against C. krusei cells and M. smegmatis on the Mendy fabric. e Mendy fabric is the most used material to make uniforms for nurses in most Zimbabwean hospitals. However, it was noted that, in most hospital setups, nurses are the most vulnerable to microbial contamination since they are the primary handlers of infected and sick people [63]. In addition, nurses spend most of the time in sick people's wards, which may increase the chances of cloth contamination by infection pathogens [63].
Microbial contamination of health workers' cloths, surfaces, and equipment plays a significant role in the spread of pathogens [15]. Successful transmission of pathogens depends on many factors, including the capacity of microorganisms to remain viable on dry surfaces, their resistance to disinfectants, and the frequency at which infected surfaces or equipment comes in contact with patients and healthcare staff [15]. Furthermore, antimicrobial resistance could be acquired through exposure of microbes to deferent conditions and substances while being attached to textiles and surfaces [64,65]. erefore, evaluation of possible alternative sources of biocidal compounds was investigated in this study.
Compound Pc6978p had 86% biocidal activity on C. krusei cells attached to the Mendy fabric. at is, at a concentration of 100 µg/mL, β-sitosterol removed and was fungicidal to about 5.04 × 10 8 C. krusei cells per cm 2 of Mendy fabric after 30 minutes of incubation. ese observations show that Pc6978p had better biocidal activity compared to the known biocide chlorhexidine digluconate which had 80% activity against the same cells under the same condition. However, there was no statistically significant difference between the biocidal activity of Pc6978p and chlorhexidine digluconate. From this study, it can be proposed that the compound can be a candidate for the formulation of alternative biocides used to decontaminate textiles. A study by Diba [66] has shown that hospital textiles, mostly blankets, are the main sources of C. krusei contamination. Several studies indicate that different Candida species can have similar adherence properties to cotton and other fabrics [66], as well as to plastics and other synthetic materials [66]. Furthermore, it is suggested that significant environmental contamination in hospital setup is from Candida species [66], especially antifungal C. krusei [67]. It is possible that the strong hydrophobicity in the C. krusei cell membrane allows stronger attachment to nonpolar surfaces and these include catheter surfaces, compromising sterility and differential diagnosis [35]. erefore, β-sitosterol can also be formulated into some of the disinfectants that are used in hospital settings to reduce C. krusei contamination. Hence, further studies on the biocidal effects of Pc6978p against C. krusei on other surfaces are recommended.
Even though compound Pc6978p showed high growth reduction effects on M. smegmatis in in vitro antimicrobial assays, it failed to exhibit biocidal activity against the M. smegmatis attached to the Mendy textile. Pc6978p only reduced 6% of the viable cells compared to the positive control chlorhexidine digluconate that had 90% biocidal activity under the same conditions. e observations may suggest that Pc6978p is less effective as a biocide against M. smegmatis. However, it is important to note that the effectiveness of biocides is influenced by many factors, such as concentration, contact time, and environmental conditions to which it is applied [68,69]. e condition, whether in suspension, adhering to a surface, or in a biofilm, would have an effect on the effectiveness of biocides [70]. erefore, all these effects should be investigated in future, so that full biocidal potential of β-sitosterol and other natural compounds from this plant can be exploited.

Conclusion
P. curatellifolia leaf extracts, extracted in this study, have been shown to contain phytochemicals that inhibit the growth of test bacterial and fungal species.
is is in agreement with the use of this plant in traditional medicine. e DCM-methanol and ethanol extracts were the most active against most test bacterial species. Mycobacterium smegmatis was the most susceptible to most extracts. β-sitosterol showed fungicidal activity against C. krusei and potent biocidal activity against C. krusei cells attached to the Mendy textile. However, further work has to be done to determine if the isolated compounds can be used to treat fungal and bacterial related ailment in model system, as well as to unlock their potential use in human health industry as biocides.
Data Availability e datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.