Murraya koenigii (L.) Sprengel seeds and pericarps in relation to their chemical profiles: new approach for multidrug resistant Acinetobacter baumannii ventilator‑associated pneumonia

Acinetobacter baumannii is without a doubt one of the most problematic bacteria causing hospital‑acquired nosoco‑ mial infections in today’s healthcare system. To solve the high prevalence of multi‑drug resistant (MDR) in A. baumannii , we investigated one of the medicinal plants traditionally used as antibacterial agent; namely Murraya koenigii (L.) Sprengel. The total methanolic extracts of seeds and pericarps were prepared and their anti‑bacterial activity was assessed using the agar diffusion method and minimum inhibitory concentration (MIC) was then calculated as compared to tigecycline. Then, an in‑vivo murine model was established which confirmed the promising activ‑ ity of M. koenigii seeds in demonstrating anti‑bacterial and anti‑inflammatory actions. The histopathological study of lungs, scoring of pulmonary lesions, counting of bacterial loads after infection by multi‑drug resistant A. baumannii all provided evidence to support these findings. LC–MS/MS profiling coupled to molecular networking and che‑ mometrics detected the presence of carbazole alkaloids, and coumarins as dominate metabolites of the active seed extracts. Positively correlated metabolites to antibacterial potential were 6‑(2 ʹ ,3 ʹ ‑dihydroxy‑3‑methylbutyl)‑ 8‑prenylumbelliferone, scopoline, and 5‑methoxymurrayatin. An in‑silico study was also performed on the crystal structure of MurF from A. baumannii (PDB ID: 4QF5), the studied structures of the mentioned extracts revealed good docking interaction at the active site suggestive of competition with the ATP ligand. These collective findings sug‑ gest that extracts of Murraya koenigii (L.) Sprengel seed is a novel prospective for the discovery of drug candidates against infections caused by MDR A. baumannii .


Introduction
Acinetobacter baumannii is a prominent and increasingly prevalent bacterium that causes severe illness and mortality [1].It is one of the most common nosocomial diseases due to its capacity to develop mechanisms of resistance to many last-line antimicrobial treatments containing carbapenems [2][3][4].In 2017, the World Health Organization (WHO) designated this bacteria as a priority-1 pathogen [5].The bacteria lead to a wide range of infections, including ventilator-associated pneumonia, soft tissue, skin, wound, and urinary tract infections [6].
Most A. baumannii infections occur in seriously ill patients in the intensive care unit setting accounting for up to 20% of infections in ICUs worldwide [7].Furthermore, the prevalence of A. baumannii infections in the community has been steadily growing.The multi-drugresistant A. baumannii (MDRAB) phenotype may invade both biotic and abiotic surfaces and form as biofilm.Numerous anti-bacterial medications, such as tigecycline, carbapenem, polymyxin, and non-antibiotic therapy, are in demand due to A. baumannii's pathogenicity [8].The detection of lactamase, low permeability of the outer membrane (OM), and effective pump systems are the primary causes of MDRAB resistance to traditional antibacterial drugs [8].
Colistin and tigecycline are antibiotics of last resort used to treat a variety of multidrug-resistant bacteria, although there have been reports of antibiotic resistance against these drugs worldwide [9].In perspectives of therapeutic strategies, herbal medicines are one of the probable approaches, which is an efficient alternative to develop several bioactive derivatives.
Curry (Murraya koenigii (L.) Sprengel) is a small aromatic shrub in the Rutaceae family [10].Phytochemical study of its pericarps and seeds extracts revealed the existence of alkaloids, flavonoids, and phenolic contents, all of which have enormous potential to improve consumer health and reduce illness risks [11].As such Murraya species can be considered as rich source of antibacterial compounds including carbazole alkaloids, and phenolics [12][13][14].
It is known that MurF is essential during peptidoglycan biosynthesis.It is an appealing target for multiple resistant bacterial treatment [15].Discovery of novel therapeutic compounds is urgently needed to overcome bacterial resistance.Therefore, an in-silico approach using the C-Docker protocol in Discovery Studio 4.0 Software can be performed on the metabolites that are positively correlated with the antibacterial activity.

Plant material
Seeds and pericarps of Murraya koenigii (L.) Spreng.(Family Rutaceae) were collected in September 2022 from Orman botanic garden (Giza, Egypt), authenticated by Mrs. Therese Labib, Botanical Specialist and Consultant at Orman and Qubba Botanical Gardens, Egypt.The voucher specimen was deposited in the herbarium of Cairo University's Pharmacognosy Department, Faculty of Pharmacy (#5.5.2022I).

Preparation of the extracts
Fresh pericarps (350 g) and 150 g of fresh seeds were airdried and extracted by methanol (3 × 5 L) by maceration for 3 days.The collected extracts were filtered and evaporated under reduced pressure to give 28, and 13 g of total methanolic extract of M. koenigii pericarps and seeds, respectively.

Chemical profiling, molecular networking and metabolites annotation
Aliquots of the samples (2 mg) were re-suspended in 1000 μL methanol: water (UPLC-grade, 1:1, v/v) and moved to the autosampler, 2 μL was injected and separated on RP High Strength Silica (HSS)T3 C18 column (100 mm × 2.1 mm having 1.7 μm diameter particles, Waters), using a Waters Acquity UPLC system.The mass spectra were obtained by full scan MS in positive ionization mode on an exact high resolution Orbitrap-type MS (Thermo-Fisher, Bremen, Germany) [22].Metabolites were identified using their mass spectra, and by comparison with our in-house database and the references literature.

Microorganisms
Extensively drug-resistant A. baumannii (XDRAB) strains clinical isolate (Additional file 1: Fig. S1) was collected from a tertiary care center in Cairo from a post covid secondary bacterial infected patient and characterized in our previous work [26].

In vitro antibacterial assay
Preliminary screening of antibacterial activity was first screened for inhibitory zone by the agar disc-diffusion method according to CLSI guidelines as described before [27,28].Furthermore, the microbroth dilution method was implemented to determine MIC (the minimum bacteria growth inhibitory drug concentration) [29].

In vivo antibacterial assay
Male Balb C mice (22 to 25 g, 8 weeks) were obtained from the Egyptian Drug Authority, and kept in a standard controlled condition.The protocol was approved by the Animal Experiment Ethics Committee of the National Hepatology & Tropical Medicine Research Institute (NHTMRI) for Research Ethics Committee (REC) (#NHTMRI REC A5-2023).All C57BL/6 mice were anesthetized by inhaling isoflurane [30].Then the mice were in the position of head up and upright, and A. baumnii suspension was dripped into the nasal cavity for 1 × 10 9 cfu in 50 μL of phosphate-buffered saline (PBS).The mice in Sham group were dripped with the same volume of normal saline by the same method.After inoculation, the mice kept their heads upright for 20 s to ensure that bacterial suspension or normal saline could enter both lungs evenly due to gravity.When the mice woke up, they were placed in the cage to eat freely.After an incubation period of 4 h, all the infected mice were randomly divided into five groups (n = 10) as follows: (1) The Sham control group (without treatment).
The Sham and model group were administrated by oral gavage with the same volume of normal saline.After 72 h of drug administration, the animals were sacrificed by cervical dislocation, lungs were collected for biochemical and histopathological examinations [31].
Pulmonary bacterial loads Lungs were removed aspetically and homogenized with a tissue homogenizer in 5 ml of sterila saline.Homogenized lung samples are then serially diluted in sterile saline and plated on Lauria bertani agar plates (Additional file 1: Fig. S2) [33].
In silico study The crystal structure of MurF from Acinetobacter baumannii complexed with ATP was successfully downloaded from Protein Data Bank (PDB ID: 4QF5) [15].The protein was cleaned.Also, hydrogen atoms was added to complete any missing residues in amino acids.It's worth noting that water molecules was removed and all unneeded molecules.Force Field CHARMm and MMFF94 partial charge was successfully applied.Protein was prepared and minimized; the active site was well defined as ATP is the main ligand.The ligand ATP was removed before docking of the tested compounds.
Statistical analysis Analysis of Variance (ANOVA) was used to establish statistical significance.The aligned peak list obtained by MZmine software and was exported as a CSV file, with information about retention time, the feature ID number, peak intensity, and mass-to-charge ratio (m/z).All variables were log 10 -transformed scaled to Pareto variance before PCA and OPLS-DA.The webbased platform Metaboanalyst 5.0 (https:// www.metab oanal yst.ca/) was utilized for Multivariate data analysis [34].

Chemical profiling
The methanolic extracts of the M. koenigii seeds (MKS) and pericarps (MKF) were analyzed using UPLC-MS/ MS (Table 1), where the mass spectra in positive modes resulted in the identification of 102 compounds viz.40 alkaloids, 35 phenolics (11 flavonoids, five phenolic acids, and 19 coumarins), five organic acids, five phospholipids, three fatty acids, ten amino acids, one quinoline derivative, one sugar, one vitamin, and one fatty amide.The mass fragmentation of those compounds was compared to reference papers as cited in Table 1.
The most prominent compounds belong to the carbazole alkaloids class; mukoline, koenimbine, murrayanine, and murrastanine A (peaks # 72, # 76, # 101, and # 102) in seeds where, koenimbine, isomahanine, 9-formyl-3-methyl carbazole, and koenine (peaks # 65, 76, # 73, # 85, and # 54) with Mexoticin (# 94) as coumarin were the prominent in pericarps.In the supplemental materials shown in Additional file 1: Figs.S3-S6, the representative MS/MS spectra of selected compounds among the major classes are shown.The molecular networking approach was applied, and this enabled the direct visual examination of MS/MS data, as well as the observation of metabolite distribution among the various extracts (Fig. 1).It categorized molecules into families or clusters based on the similarity of their MS/MS fragmentation patterns.

In vitro anti-A. baumannii study
Upon performing antibacterial susceptibility by Kirby's Bauer disc diffusion, M. koenigii seed extracts showed high antibacterial activity (21 mm) and could diminish bacterial growth with A. baumannii being sensitive towards the sample as well as tigecycline (20 mm).By contrast, M. koenigii fruit extracts were ineffective with inhibition zones of 10 mm (Table 2 and Additional file 1: Fig. S7).Furthermore, MICs for M. koenigii seeds and fruit extracts were 32, and 125 ug/ml, respectively (Table 2).

In vivo anti-A. baumannii study Hematological biomarkers
M. koenigii seed extracts and tigecycline envoked few changes in regard to these parameters.By contrast, the fruit extract showed higher levels of hematological biomarkers revealing higher inflammation and lower antibacterial activity (Fig. 2).

Histopathology
As illustrated in Fig. 3 and Table 3, no histopathological changes were observed in lung sections from normal group as normal bronchioles and alveoli were observed.By contrast, the untreated group exhibited severe diffuse pneumonia manifested by heavy inflammatory cells infiltrations that masked lung tissue with discreet area of pulmonary necrosis.The alveolar lumens were heavily infiltrated by inflammatory cells with presence of bacterial colonies.Blood vessels with severely congested.M. koenigii seeds extract treated mice showed marked improvement as most of the studied sections demonstrated mild interstitial mononuclear inflammatory cells infiltrations with clear alveolar lumen and mild vascular congestion.The M. koenigii pericarps extract treated group exhibited mild improvement as the examined tissue samples showed moderate inflammatory cells infiltrations in the interstitial tissue as well as multifocal areas of pneumonia.Tigecycline S/C group showed marked improvement as most of the examined sections were apparently normal, only a few sections exhibited mild interstitial infiltration with mononuclear inflammatory cells.

Pulmonary bacterial loads
M. koenigii seeds showed better anti A. baumannii than pericarps which had a moderate activity as revealed from bacterial loads results (Table 4).

Multivariate data analysis
Untargeted metabolomic analyses usually generate a complex dataset in terms of features and their corresponding intensities, thus, various dimensionality reduction methods including principal component analysis (PCA) and hierarchical cluster analysis (HCA) are performed to ease the process of visualizing the data.PCA The results of unsupervised analysis was further verified using the supervised orthogonal partial least squares discriminant analysis (OPLS-DA), Offering valuable insights into the distinguishing metabolites between the tested samples.The OPLS-DA score plot (Fig. 5A) explained 65.9% of the total variance and 15.3 of the orthogonal total variances, where seeds and pericarps segregated into two different nonoverlapping clusters.The loadings S-plot (Fig. 5B) was used to compare variable magnitude vs reliability, with axes displayed from the predictive component being the covariance P [1] versus the correlation P(cor) [1].Isogirinimbine, scopoline, sinapine, isomahanine, mukoline, Murrastinine B, 8, 8″-bis koenigine, N-Methylaniline, sinensetin, and 5-Methoxymurrayatin were the discriminating metabolites of seeds.Furthermore, quercetin, isoquercitrin, toddalenone, bismurrayafoline A, norrangiformic acid, meranzin hydrate, 8,8'-Bismurrayamine A, dimethyl ether, Murrastinine C/ murrayacine, koenigine, and mexoticin were the discriminating metabolites in pericarps.
In addition, the metabolites that distinguished between active and inactive extracts were further validated by computing the VIP scores obtained from the OPLS-DA modelling of the active seed extract against the inactive fruit extract.As can be concluded, the discriminating metabolites were relevant to explain the variance when also having VIP scores > 1at p < 0.05 (Fig. 5C).The metabolites positively correlated to the antibacterial activity are shown in Table 5.

In silico study
The tested ligands were prepared, then docked into the binding site of MurF (A. baumannii).The binding mode of the tested compounds was analyzed and compared to that of the ligand to investigate the antibacterial activity.Ten compounds showed best results regarding the binding mode to the key amino acids in the active site, compared to the ATP ligand.The C-Docker results of the ATP ligand (E = − 142.63 kcal/mol) was used to evaluate the docking interaction of the test compounds as shown in Fig. 6.Where the main reported binding interactions including Hydrogen Bond Acceptor interaction (HBA) with Ser123, Hydrogen Bond Donor (HBD) interaction with His292, and HBA interaction with Asn296.In addition to HBD with Asp341 and HBA with Lys448 in the central domain was accomplished.While in the C-terminal domain, pi donor hydrogen bond interaction with Ser349 and hydrophobic interaction with Ala352 was noticed.Moreover, charge interactions with Lys125, and Arg327 in the central domain was confirmed [15].In addition to, 3HBA with Asn344, 1 HBA withThr126, Thr127, Asn150 via the phosphate groups of ATP.And 1 HBD with Gln295 and pi donor hydrogen bond with Thr348.
While the two bulky compounds named: 8,8ʺ-biskoenigine and murrafoline A, failed to interact within the active site and showed no docking results.interaction energy (Fig. 8a).The visual inspection of the binding mode of the tested compounds confirmed the essential binding interactions with the reported amino acid residues at the active site compared to ATP ligand (Fig. 8b).

Discussion
Murraya koenigii (L.) Sprengel or the Curry plant is a potential medicinal plant, which belongs to the family Rutaceae, highly known for its characteristic aroma and bioactive compounds.Phytochemical analysis of the plant has revealed the presence of proteins, carbohydrates, vitamins, alkaloids, phenolics, and flavonoids providing enormous possibilities to improve consumer health and reduce illness risks [11].Carbazole alkaloids such as koenine, mukoeic acid, mahanimbine, koenimbine, murrayazolidine, murrayazoline, murrayacine and girinimbine have been identified as biologically active compounds with antioxidant, antimicrobial, anti-inflammatory, anthelmintic, antidiarrheal, hepatoprotective, analgesic, and cytotoxic properties [12].Carbazoles such as mahanimbine, girinimbine, koenimbine, isomahanine and mahanine were isolated from the pulp of pericarps and seeds of M. koenigii.Coumarins were isolated and characterized from the seeds [12,35,36], as were a wide range of phospholipids and fatty acids [37].In our investigation of extracts from the seeds and pericarps of M. koenigii, LC/MS/MS analysis allowed the identification and quantification of 40 carbazole alkaloids, 19 coumarins, 11 flavonoids, and five phenolic acids.
Murraya koenigii carbazole alkaloids and coumarins have been demonstrated to be antibacterial, with even greater efficacy than the antibiotics Amikacin and Gentamicin against Staphylococcus, Streptococcus, Escherichia coli, Pseudomonas, Klebsiella, and Proteus sp.[38].So, the presence of some coumarins such as murrayanone, and scopoline and carbazole alkaloids as mukonine, and iso-koenigine demonstrates the potential of seeds than pericarps as anti A. baumannii in our study.Plants of the family Rutaceae are widely used in various parts against Gram-positive and Gram-negative harmful bacterial strains as extracts from this family are dominated by the two important bioactive classes of compounds namely the coumarins and carbazole alkaloids [39,40].In-vitro and in-vivo studies revealed promising activity of the seed extracts which were almost as active as the standard Tigecycline.Multidrug-resistant A. baumannii is a pathogen that causes severe infections in critically ill people and is infamous for propagating epidemically [41].By producing pro-inflammatory cytokines, the host activates the innate and adaptive immune systems.However, because over-activation of pro-inflammatory cytokines leads to multi-organ failure, macrophages' early   with A. baumannii strains.IL-10 has also been implicated in vaccine-induced protection against A. baumannii challenge, where pro-inflammatory cytokines (TNF-and IL-6) levels were significantly reduced and anti-inflammatory cytokine IL-10 levels were significantly increased in lungs and serum, resulting in decreased severity and slow progression of disease [46].These protective effects were reflected in the hematological parameters measured, and from the histopathological study also where the treatments alleviated the inflammatory cells infiltration, pulmonary necrosis, and vascular congestion.In addition to the reduction of pulmonary bacterial loads and decreasing the high mortality rates after infection with highly resistant A. baumannii strain.The majority of the positively linked biomarker metabolites have been shown to have antibacterial action against distinct strains of bacteria including sinapine [47], sinensetin [48], quercitrin [49], ferulic acid inhibits tetK and MsrA efflux pumps of multidrug resistance strains [50] in addition to scopolin which is a coumarin derivative that could inhibit the activity of p-glycoprotein (p-gp) and other multidrug resistance proteins [51].Validation was confirmed by re-docking of the ligand in the target active site showing RMSD value = 0.5A° as good validation result.It was reported from the previous literature [52] that the binding mode of ATP at the active site is mainly concerned with the interface between the Central and the C-terminal domains of MurF, showing several interactions with amino acid residues.Where MurF structure is mainly composed of three domains named: the N-terminal, the Central, and the C-terminal domains, respectively [15].It is known that MurF is essential during the last step in the biosynthesis of monomeric precursor of peptidoglycan within peptidoglycan biosynthesis since it adds (in an ATPdependent manner) D-Ala-D-Ala dipeptide to UDP-Nacetylmuramyl-L-Ala-γ-D-Glu-m-DAP.What makes MurF became an attractive target to develop novel antibiotics.
Here, we used computer added drug design (CADD) applying molecular simulation via working on the crystal structure of the Acinetobacter baumannii MurF (AbMurF)-ATP complex [15].The molecular docking study of the tested compounds revealed twelve promising compounds with comparable and competitive binding mode to ATP as ligand in the binding site.Furthermore, the potential of cabazole alkaloids as antibacterial against multidrug resistant strains was reported several potential actions helping to solve the MDR problems worldwide [19].Therefore, in order to gain insight to the potential of positively correlated carbazoles including isogirinimbine, 8, 8''-bis koenigine, mukonine, murrastinine B, iso-Koenigine, mahanimbidine, and murrafoline, a further detailed studies of these compounds are required.It is worth highlighting the prophylactic efficacy against A. baumannii infection in a murine model of Murraya koenigii (L.) Sprengel seeds and pericarps is reported here for the first time.
Trying to deal with the unmanageable multidrug resistance towards the existing antibacterial agents, novel approaches towards the discovery of new targets are required in order to prevent and treat A. baumannii infections.While clinical monitoring, developed antibiotics have been successfully targeting enzymes involved in the synthesis of peptidoglycan.

Fig. 1 A
Fig. 1 A Molecular networking established using LC-MS/MS data in the positive ESI mode.B Selected nodes and clusters have been zoomed in.Single nodes are used to represent molecules that do not form groups. Nodes were also colored by organ type (seeds and pericarps) and labelled with their precursor m/z values.The nodes were also represented as a pie chart to show the semi-relative abundance of the identified molecule ions, with the borders indicating the mass differences between the connected nodes

Fig. 2
Fig. 2 Hematological biomarkers of seeds and pericarps of M. koenigii as compared to Tigecycline in Acinetobacter baumannii murine animal model.Data are expressed as mean ± SD.Statistical analysis was carried out by one-way ANOVA followed by Tukey's multiple comparison test.a Significant difference from normal group at p < 0.05.b Significant difference from infected group at p < 0.05.ab Significant difference from normal group and infected group at p < 0.05

Fig. 4 A
Fig. 4 A PCA and B HCA loadings plot.Multivariate data analysis was performed with the complete LC-MS dataset

Fig. 5 A
Fig. 5 A PLS-DA, B S-Plot and C VIP Top 25 correlated with the antibacterial activity

Table 2
Screening of antibacterial activity of Murraya koenigii seeds and pericarps against A. baumannii R resistant, S sensitive

Table 5
Metabolites dominate the active Murraya koenigii seeds extract