Antibacterial Activities of Cationic Porphyrins and Porphyrin Encapsulated Gold Nanorods on Bacterial Cell Lines

Diseases caused by bacteria, parasites and fungi have become a major problem globally, this is despite the exceptional advances in medicine that have been reported thus far 1-3. The impact of these diseases is very panoptical in developing and underdeveloped countries due to the lack of modern medicine in these countries. Resistant antimicrobial pathogens are increasing progressively, hence the need for the current antimicrobials to be reviewed and renewed to improve effectiveness 4-8. Advances have been made by the development and conjugation of new organic molecules with metal nanoparticles over the years for applications against resistant, and multiple drug-resistant bacteria.


INTRODUCTION
Diseases caused by bacteria, parasites and fungi have become a major problem globally, this is despite the exceptional advances in medicine that have been reported thus far 1-3 . The impact of these diseases is very panoptical in developing and underdeveloped countries due to the lack of modern medicine in these countries. Resistant antimicrobial pathogens are increasing progressively, hence the need for the current antimicrobials to be reviewed and renewed to improve effectiveness [4][5][6][7][8] . Advances have been made by the development and conjugation of new organic molecules with metal nanoparticles over the years for applications against resistant, and multiple drug-resistant bacteria.
The use of antimicrobial photodynamic therapy (aPDT) as a treatment modality entails the use of photosensitizers (PSs) and visible light for the induction of oxidative damage to the microbial pathogens including bacteria 9,10,11,12 . The method can be used as an alternative to traditional antimicrobial medicines in the treatment of localized infections. Reports have shown the use of cationic porphyrins to dominate anionic or neutral porphyrins in terms of antimicrobial activity [13][14][15][16] . The presence of chlorine, fluorine or nitro groups not only improves the polarity and lipophilicity of the photosensitizers but also improves their antitumor activity [17][18][19][20][21] . The addition of a central metal atom in the porphyrin structure also improves the porphyrin polarity and lipophilicity 22 . Figure 1 shows a simplified method for the synthesis of the desired cationic porphyrins and encapsulation of AuNRs.
Metal nanoparticles have been utilized since ancient times in the fight against pathogenic bacteria [23][24][25] . Due to simple synthesis, their distinctive photophysical properties, and their biocompatibility, gold nanoparticles (AuNPs) are used in various biological applications. Their surface plasmon resonance (SPR) has opened doors for the particles to be used in heating, sensing, imaging and drug delivery [26][27][28] . The synthesis of the nanoparticle can also be tuned to give the desired shape to improve the SPR frequency of absorption when irradiated with nearinfrared (NIR) light 29,30 . The light energy absorbed is converted to heat, generating higher temperature around the environment in which the nanoparticles are embedded. AuNRs can absorb and dissipate light in this way over a wider range of frequencies, compared to other shapes of nanoparticles; from the visible to the near infra-red, depending on their aspect ratio, and this makes them ideal candidates for photo-thermal antibacterial applications [31][32][33] .
The interest of this study is on two cationic porphyrins, TPyP and TAP, how they are used in the encapsulation of AuNRs, and their antibacterial activity against four drug-resistant bacteria without using light. Furthermore, molecular docking studies have become very popular in the field of drug discovery and design as this has been used as a tool to further understand how the ligands interact with specific protein targets. The second purpose of the research is therefore to explore the interactions of the two porphyrin compounds with the bacterial DNAgyrase enzyme, by using molecular docking studies using the Biovia Studio 4.1 visualizer, with the interaction with ciprofloxacin used as the control. Also, the SwissADME online tool was used to assess the pharmacokinetic properties of the compounds, focusing on bioavailability and the potential to cross the blood-brain-barrier. The toxicological properties were investigated using the online server ADMETlab online tool based on the potential to inhibit the human ether-a-go-go-related (hERG) gene, the HHT and mutagenicity. To the best of our knowledge, the antibacterial activities of AuNRs encapsulated with these porphyrins have not been studied hence the study. These studies will provide the baseline antibacterial activity of these nano-conjugates without the use of light. Additionally, the computer simulations of the bacterial DNA-gyrase interactions of these nano-conjugates will indicate the topoisomerase potential in line with the observed aPDT activity, while the ADMETlab online tool toxicological simulations will indicate the possible toxicological impacts, for further investigation.

MATERIALS AND METHODS
Reagents were purchased from Sigma-Aldrich (Steinem, Germany). The chemicals were used without any purifications except the pyrrole which was distilled before use. The bacterial strains used for this study included the Gram-positive bacteria, E. faecalis (ATCC#29212), S. aureus (ATCC# 25925) and the Gram-negative bacteria K. pneumonia (ATCC#4352) obtained from the microbiology culture bank at the University of Zululand where all the microbiological tests were carried out. A detailed description of the synthesis of the two porphyrins; TPyP and TAP, their encapsulation and characterizations are given below.

Synthesis of 5,10,15,20-tetrakis(4-pyridyl)porphyrin (TPyP)
The preparation was conducted as described in the literature using the Adler method (33) with modifications. A mixture of pyridine carboxaldehyde (4.71ml, 50.0 mmol) and acetic anhydride (6.50ml, 68.8mmol) was dissolved in 50 ml propionic acid and freshly distilled pyrrole (3.24ml, 50.0mmol) in 10ml propionic acid and added dropwise while refluxing under N 2 atmosphere for 2-4 hours. The reaction mixture was allowed to cool down overnight in open air and propionic acid was removed by distillation at 141 º C obtaining dark purple precipitate. The precipitate was washed with hot water followed by the neutralization with aqueous ammonia and further washed with hot water to obtain a purple solid. The purple solid was further treated with dichloromethane, followed by vacuum filtration obtaining TPyP which was dissolved in chloroform and purified by column chromatography using chloroform, hexane and methanol as an eluent to obtain purple crystals 34  Synthesis of 5,10,15,20-tetrakis(4-nitrophenyl)porphyrin 5,10,15,20-tetrakis(p-nitrophenyl)porphyrin (TAP) was synthesised using the Lindsey reaction as in literature. P-nitro benzaldehyde (34 mmol, 5.26 g) and acetic anhydride (64 mmol, 6 ml) was added to 50ml propionic acid while stirring under a nitrogen atmosphere and the resulting solution was heated to reflux. Freshly distilled pyrrole (34mmol,11.5ml) in 10ml propionic acid was added to the refluxing solution and allowed to reflux for a further 30 minutes while stirring. The resulting reaction mixture was allowed to cool to room temperature and for a further 24 hours in open air. The black solution was filtered under vacuum and a black tarry powder was obtained. The black tarry powder was transferred to a 150 mL round-bottomed flask, 40ml of pyridine was added and the mixture was refluxed for an hour. The mixture was cooled to room temperature and further cooled at -40 0 C for 24 hours. The tarry liquid was filtered and washed with acetone until the rinsing was no longer dark to yield 5,10,15,20-tetrakis(4nitrophenyl)porphyrin.
Synthesis of 5,10,15,20-tetrakis(4-aminophenyl)porphyrin (TAP) 5,10,15,20-tetrakis(nitrophenyl)porphyrin (2.5 mmol, 2 g) was bubbled for one hour in 100 ml HCl (0.1 M) with N 2 . Stannic chloride hydrate (SnCl 2 .H 2 O: 9.0128 g, 40 mmol) in 10 ml HCl (0.1 M) was bubbled with N 2 and added to the solution. The resulting mixture was then heated in a water bath at 78 0 C for 30 minutes and then slowly transferred to a coldwater bath and an ice bath at -4 0 C. Aqueous sodium hydroxide (0.1M) was added to the reaction mixture to neutralize the solution under cold temperatures. The solution was then filtered to obtain a greenish solid which was stirred vigorously with sodium hydroxide (200 ml: 0.1 M). The resulting solid was then filtered, dried and purified using a Soxhlet extractor in chloroform to give purple crystals. Yield (1.97 g, (21%). IR

Resuscitation of Microorganisms
The bacteria selected were inoculated in a nutrient broth and incubated at 37 °C overnight. Following this, 1.0 ml of the bacteria in the nutrient broth was then pipetted in 9ml freshly prepared nutrient broth placed in separate test tubes labelled with corresponding microorganisms. The test tubes were also incubated overnight at 37 °C. The next step was to read the absorbance of the selected bacterial strains on the spectrophotometer at 600nm to determine their turbidity. The turbid resulting suspensions were diluted with nutrient broth to obtain an absorbance of 0.132 nm which is comparable to that of the 0.5 McFarland turbidity standard taken to consist of 1.5 x 10 -3 colony forming units (CFU)/ml 37 .

Minimum Inhibitory Concentration (MIC)
The test compounds were investigated for antibacterial activity against S. aureus (ATCC 25925), E. faecalis (ATCC 29212) and K. pneumoniae (ATCC 4352) using the micro-dilution method. Mueller Hinton broth (MH) (100 μl) was pipetted into each of the wells of a 96-well micro-plate. The test compounds were then added into the first row of the micro-plate and serially diluted to a range between 50 and 0.313 mg/ml. Thereafter, 100 μl of the fresh bacterial cultures at a density of 1 x 10 6 CFU·ml -1 , was pipetted into the wells. DMSO (10%) was used as negative control while ciprofloxacin (20 μg/ml) served as a positive control. The micro-plates were sealed and incubated at 37 °C overnight. Afterwards, p-iodonitro tetrazolium violet solution (INT: 40 μl of a 0.2 mg/ml solution) was added to each of the wells and the micro-well plate was incubated at 37 °C for 30 minutes. The MIC was perceived as the lowest concentration of the test compounds to inhibit bacterial growth 38 . Figure 2 shows a simplified procedure for the determination of the MIC and the MBC.

Minimum Bactericidal Concentration (MBC)
MBC of the test compounds was assessed by removing a loopful of each culture from the wells that had no bacterial growth and streaking it on sterile MH agar plates. The agar plates were incubated at 37 °C for 24 hours. The lowest concentration of the test compounds that exhibited the complete killing of the tested bacterial strains was considered as the MBC 39 .

Synthesis of TPyP and TAP UV/Vis Spectroscopy
Synthesis of TAP and TPyP was studied using the Ultraviolet and Visible spectroscopy (UV-Vis). The spectrum of TPyP and TAP show two band sets; one Soret band and four Q bands. The soret band appears at 512nm while the Q bands are at 412, 512, 546, 586 and 645 nm for TPyP and 427, 526, 565, 598, and 660 nm for TAP. The shape of the bands is of phyllo type due to the amino-phenyl groups attached at meso positions.

FTIR Spectroscopy
The FTIR spectrum of TPyP synthesized at different times is in total agreement with the structure. Absorptions at 3372 and 883 cm -1 are a result of the stretching and bending vibrations of N-H and C-N respectively, this is a characteristic of a free base porphyrin. The intense band that is observed at 760 cm -1 is attributed to the C-H bond from the pyrrolic ring. At 671 cm -1, the band observed is of the C-H sp 2 . At 1209 cm -1 there is a band that signifies the N-H out of phase bending while the 2972 cm -1 signifies one of the C-H pyrrolic rings, sp 3 . The functional groups of TAP were confirmed and studied using a Fourier Transform Infrared Spectrometer. The spectrum obtained for the TAP was in agreement with the expected structure. The band at 748 cm -1 is a result of C-N stretching and bending vibrations characteristic of the free-base porphyrins. The band at 2976 -1 is the N-H out of phase and is more significant in the porphyrin synthesized in an ionic liquid than in propionic acid.

NMR Spectroscopy
The structure of TPyP was confirmed using 1 H 600MHz NMR spectroscopy. At 8.87 ppm lies a doublet from the two β protons. The singlet at 8.89 ppm is attributed to the 8 protons meta to N on the pyridyl ring. The doublet at 8.19 ppm is due to the protons ortho the N in the pyridyl ring. Lastly, at -2.90 ppm a singlet is observed attributed to the protons on the NH on the interior of the porphyrin ring. The confirmation of the structure of TAP was done using 1 H 600MHz NMR spectroscopy. The peaks coincided well with the structure. The singlet at 8.87ppm is due to the β pyrrole protons. At 8.26ppm, a doublet is seen attributed to the 8 protons meta to the NH 2. The doublet of doublets at 7.05 ppm is due to the 8 protons ortho to NH 2 . The signal at -2.75ppm is attributed to the protons at the NH on the interior of the porphyrin ring.

Antibacterial evaluation
The microdilution method was used to investigate the minimum inhibition concentration (MIC) of the test compounds. The test compounds showed broad-spectrum activity against all tested bacterial strains. The minimum inhibitory concentration of the test compounds ranged from 0.78 mg/ml to 2.60 mg/ml for TPyP-AuNRs and 0.42-1.8mg/ml for TAP-AuNRs as in Table 1. The effect of the test compounds was more pronounced against Gram-positive bacterial strains than Gram-negative strains. Cationic porphyrins effectively interact with the cell walls of Gram-negative and positive and hence they photo inactivate the bacteria. A direct mechanism is not known for the interaction but the porphyrins are found to undergo Type 1 and Type II photosensitization 40 . Gram-negative bacteria (K. pneumoniae) were resistant to lower concentrations of the test material (x<50mg/ml). This could be attributed to the thin peptidoglycan layer of the outer membrane as seen in Figure 3. This outer membrane consists of an additional protective lipopolysaccharide layer that exhibits resistance and antigenicity against antibacterial agents and hence a conclusion was drawn that the high resistance of K. pneumoniae is due to this layer. The antibacterial activity of the test materials was due to the abiotic and biotic elicitation. Because porphyrins can inhibit the growth of the micro-organism, the highest concentration gives the best results (50mg/ml).
Generally, bacterial growth is composed of peptones (peptides and amino acids) and these are obtained through the hydrolysis of natural products or enzymatic digestion. The products can sometimes induce the aggregation of the AuNPs. The use of PEG however enhanced the disaggregation and stability of the AuNRs and also prevents nonspecific protein adsorption. MIC and MBC values of the test materials consisting of AuNRs capped with porphyrin containing PEG confirmed the bacteriostatic properties but not the bactericidal properties of the materials against the bacterial strains. Capping off the AuNRs with PEG and porphyrins reduces the aggregation of the particles. Furthermore, this enhances the particles to be more compatible with the biological media and this is due to the reduced non-specific protein absorption, and the biocompatibility of the PEG outer shell. This means that the higher the concentration, the higher the bacterial inhibition. Thus, our test materials have moderate to weak antibacterial activity. The test materials did not show any bactericidal effect against the tested strains (Table 1). The imposition of antibacterial materials on exponentially growing bacterial strains is either bactericidal, bacteriostatic or bacteriolytic. In the absence of light, therefore, the test compounds used, possess only the bacteriostatic effect and not the bactericidal effect against the tested strains as shown by the MIC and MBC data shown in Table 1.

Preparation of Ligands
The three-dimensional (3D) structures of the two ligands and the control (ciprofloxacin) were procured from the ChemSpider compound database. The compounds were downloaded in Structure-Data File (SDF) format and converted to PDB coordinates using Biovia Discovery Studio 4.1 visualizer. They were then loaded to AutoDock Vina and converted to PDBQT files 41 .

Receptor preparation
The 3D structures of the receptor protein DNA gyrase (PDB ID: 5Z9N) were obtained from the Protein Data bank (www.rcsb.org). DNA gyrase is responsible for DNA replication. Thus, DNA gyrase is one of the main targets during the process of discovery and development of antibacterial agents 42 . The receptor protein was loaded into the Biovia Discovery Studio 4.1 Visualizer. All the water molecules, heteroatoms and other ligands were removed from the crystal structure. Thereafter, the polar hydrogen was added (50). The prepared receptor was then loaded into AutoDock Vina in the PDB file and then converted to the PDBQT file to enable the docking process 41 .

Docking
The ligands were docked to DNA gyrase subunit A (PDB code: 5Z9N) using AutoDock Vina 41 . The docking simulations between the ligands and the receptor protease were performed by using the AutoDock Vina.

Pharmacokinetic properties
The SwissADME online tool was used to assess the pharmacokinetic properties of the compounds. The simplified molecular-input line-entry system (SMILES) for each compound were procured from PubChem and incorporated into the SwissADME online tool. Thereafter, the ADME properties that include absorption: gastrointestinal absorption (GA), distribution: blood-brain barrier (BBB) permeability, metabolism: isomers of cytochrome P450 (CYP) inhibitors, excretion: P-glycoprotein (P-gp) substrate were computed 43 .

Toxicological study of the compounds
The toxicological properties of the two porphyrin compounds were investigated using the online server ADMETlab online tool. The SMILES of the compounds were obtained from PubChem and inserted into the ADMETlab online tool. Thereafter, each compound was evaluated for its potential to inhibit the human ether-a-go-go-related (hERG) gene and for its human hepatotoxicity (HHT) and mutagenicity 44 .
Molecular docking is a computational tool primarily used to predict the structural orientations, binding affinities and receptor-ligand interactions 45 . To understand the mechanism of antibacterial activity of the ligands, AutoDock Vina was employed to predict the strength of binding affinity between the receptor protein and ligands using scoring functions and the results are displayed in Table 2. TPyP has a binding free energy of -8.7 kcal/mol whereas TAP has a binding free energy of -8.3 kcal/mol. It was also noted that both ligands have binding free energies lower than that of the control, ciprofloxacin (-6.9 kcal/mol).
Receptor-ligand interactions occur spontaneously when the free energy change is negative; the negative energy values indicate the stability of the receptor-ligand complexes and the lower the binding free energy, the better the binding affinity 41 . The compounds showed better binding values than the control and the results correlate with those obtained from in vitro antibacterial activity.
The free binding energy values of the receptor-ligand complexes are due to various types of molecular bonds formed. The molecular receptorligand interactions at the active site of DNA gyrase are summarized in Table 2, Figure 4  The binding forces formed are van der Waals forces, pi-sigma bonds, alkyl and pi-alkyl bonds. It was thus concluded that the ligands interact strongly with the functional groups of the DNA gyrase, suggesting potential for bacterial DNA gyrase based photodynamic inactivation. Given the topoisomerase role of the bacterial DNA gyrase enzyme, the results of these docking studies suggest that the mechanism of the demonstrated bacterial inactivation by TPyP and TAP might involve abolishing the activity of the bacterial DNA gyrase enzyme by a photodynamic mechanism.

Pharmacokinetic properties of the compounds
To be considered ideal for use as antimicrobial agents, test compounds should show high efficacy coupled with desirable pharmacokinetic properties. SwissADME online tool was used to assess the pharmacokinetic properties of the compounds. The results of the SwissADME online tool modelling studies of the pharmacokinetic properties of the test compounds are summarized in Tables 3. GA is the level of the absorption of the orally administered test compounds from the gastrointestinal tract into the bloodstream 46 . TPyP showed a high potential to be absorbed in the gastrointestinal tract whereas TAP demonstrated lower potential (Table 3). GA greatly influences the bioavailability of the orally taken compounds 45 . Thus, only TPyP is likely to be bioavailable in the body system upon oral administration, whereas TAP is likely to be poorly delivered systemically via oral administration. For these reasons, it might be more feasible to opt for the intravenous injection method for both compounds.
The blood-brain barrier (BBB) is an important factor for test compounds that target the brain cells. However, if the compounds are not targeting brain cells, their inability to cross BBB is of an advantage as they cannot induce any adverse side effects in the brain 47 . The ability for the compounds to penetrate through BBB was predicted and the results are shown in Table 3. Both compounds did not show the potential to cross A B C   BBB. The results are indicative of the inability of these compounds to be used to treat brain-related disorders and potential biosafety, as they cannot induce adverse side effects in brain cells. The broad substrate specificity ATP-dependent efflux pump P-gp plays a vital role in the trans-membrane transportation of molecules. The efflux action of P-gp tends to resist drug efficacies in the cells as it pumps the administered drugs out of the cells, consequently lowering their concentrations and effectiveness 48 . The test compounds are predicted to be the substrates of P-gp (Table 3). This implies that their pharmacological activities can be compromised by the efflux action of P-gp, clearing the cell interior and therefore forcing their activity to be on the outer cell wall. Cytochrome P450 (CYP) are heme-proteins found in the liver and are the major drug metabolising proteins. They are responsible for the metabolism of more than 75% of the currently available drugs in the market 49 . The computed effect of the compounds on the five major CYP isoforms (CYP 1A2, CYP 2C19, CYP 2C9, CYP 2D6 and CYP 3A4) are shown in Table 3. Out of the five CYP isomers, TPyP showed the ability to inhibit CYP 1A2 and CYP 3A4 whereas TAP was predicted to only inhibit CYP 1A2. The blockage of CYP isoforms translates to biodegradation and bioavailability reduction of the compounds. Moreover, the accumulation of compounds due to the inability to be metabolised can lead to adverse toxicological outcomes 50 . Nevertheless, the majority of the tested isoforms were not inhibited by TPyP or TAP, thus increasing their bioavailability.

Toxicity studies of the compounds
The majority of pharmacologically active compounds often fail in the last stages of drug discovery and development due to toxic effects such as cardiotoxicity (hERG inhibition), hepatotoxicity and mutagenicity.
The hERG is the gene in the heart muscle that encodes for the voltagegated potassium channel, which is responsible for the formation of the cardiac activity potential. It simply regulates the heartbeat by controlling the potassium current (IKr). Thus, the hERG inhibitors tend to induce cardiac arrhythmia and sudden death 45 . The effect of the compounds on hERG was evaluated and the results are shown in Table  3. Both compounds were found to have the potential to inhibit hERG. This implies that these compounds fall in the category of drugs like terfenadine and astemizole, which can induce prolonged QT-interval and heart failure 51 .
The human liver is the main site for drug metabolism. This means that it is highly vulnerable to the harmful effects of xenobiotic agents and drugs. Human hepatotoxicity (HHT) includes any form of injury to the liver that may result in organ failure 51 . The HHT of the compounds were predicted and the results are illustrated in Table 4. The compounds displayed no human hepatotoxic effects. This implies that the compounds cannot induce any impairment in the liver and cannot temper with its functions. Ames test is used to evaluate the potential mutagenicity of compounds. The mutagenic compounds can cause mutations in the genes and cause cancerous cells 52 . The mutagenic effects of the compounds are shown in Table 4. The compounds showed the potential to induce mutagenic effects in humans. This means the compounds pose a great threat to human health as they have the potential to generate cancer cells in the body.    compounds did not show bactericidal effects but only bacteriostatic, they can be used as potential agents against bacterial infections and tumours. TPyP had binding free energy of -8.7 kcal/mol whereas TPA had binding free energy of -8.3 kcal/mol. They showed promising pharmacokinetic properties. Both compounds have the potential to cause cardiotoxic and mutagenic effects but not hepatotoxic effects. These findings suggest that the porphyrins TPYP and TAP could be investigated further as agents for the combination of photodynamic therapy and photothermal therapy, to establish whether there is an enhancement of their anticancer and antibacterial activity because such additive and/or synergistic enhancement would mean that the effective therapeutic doses would be much less than the MIC and MBC range observed in this study. Pharmacokinetic studies suggest that it might be more feasible to opt for the intravenous injection method for both materials.

CONCLUSION
Dr TS. Maliehe holds a BSc degree in Biotechnology and BSc-Honours, MSc and PhD in Microbiology from the University of Zululand. He is a postdoctoral fellow in the same institution. His research expertise falls within Environmental Microbiology with an emphasis on wastewater treatment using microbial bioflocculants. He is an expert in bioactive compounds of health importance from microbial origin (endophytes) and plants. He is also well informed in Bioinformatics. He has collaborated with a large number of researchers and academics and published his work in highly rated accredited journals.