Antimicrobial effects of craniopharyngioma cystic fluid

Tumours are known to increase the risk of infections, especially those occurring in the central nervous system where insertion of surgical hardware/shunts such as in craniopharyngiomas may be required. However, infections are surprisingly scarce in craniopharyngioma cases. In this study, we explored the possibility of antimicrobial effects of craniopharyngioma cystic fluid. The antibacterial effect of craniopharyngioma cystic fluid samples against selected human pathogens: Escherichia. coli, S. aureus and S. pneumoniae were determined using the agar disc diffusion method. Streptomycin and ampicillin were used as controls. The test organisms were cultured in Mueller-Hinton broth overnight at 37 °C. McFarland standard was used as a reference to adjust the inoculum size of each test organism to a concentration of 1 × 106 CFU/ml using sterile broth. The craniopharyngioma cystic fluid inhibited growth of Gram-positive bacteria S. aureus and S. pneumoniae, but not the Gram-negative bacteria, E. coli. The samples showed the highest zones of S. pneumoniae growth inhibition of up to 20.0 ± 1.0 mm compared with 18.0 ± 1.0 mm of streptomycin and 9.0 ± 0.0 mm of ampicillin. Craniopharyngioma cystic fluid showed significant antibacterial properties against Gram-positive bacteria. This novel finding has implications in the way we view infections in craniopharyngioma patients. More studies need to be carried out to further elucidate this unique finding and possibly exploit these antimicrobial properties.


Background
Craniopharyngiomas are rare benign tumours which account for 6-9% of all childhood tumours and are the most common non-glial tumours in childhood [1]. According to the World Health Organization, they account for 1.2 to 4.6% of all brain tumours [1]. There are two types of craniopharyngiomas, namely the adamantinomatous type and the papillary type. The adamantinomatous craniopharyngioma type usually occurs in childhood whilst the papillary type occurs in middleaged adults [2]. Craniopharyngiomas usually occur in the supra sellar region [3]. They are of epithelial histology and often exhibit calcification and sometimes single large or multiple cysts that contain fluid. This fluid is yellow, viscous, proteinaceous and rich in cholesterol crystals [4]. Literature has scanty information on craniopharyngiomas, let alone the tumour cystic fluid thereof [1][2][3][4].
Recently, some proteomic studies have looked closely at the molecular composition of craniopharyngioma machinery fluid. Some of the identified proteins were grouped as participating in inflammatory response (alpha defensins, beta thymosins and apolipoprotein family) [5]. Defensins have for a long time been recognized as natural antimicrobial peptides and possess diverse versatile immune functions. They possess both inflammatory and antiinflammatory properties. These properties are cell type and context specific [6]. Biological effects of these properties have not been explored in the context of craniopharyngiomas. Treatment of these slow growing tumours includes partial to total surgical resection, radiation therapy, insertion of ventriculoperitoneal shunts in the event of hydrocephalus and insertion of an ommaya reservoir to allow intermittent drainage of cystic fluid. This may involve injection of sclerosing agents like bleomycin into the cystic cavity via the ommaya reservoir [1,7,8].
Tumours are known to increase the risk of infections, possibly more so within central nervous system tumours such as craniopharyngiomas, which may require the insertion of a ventriculoperitoneal shunt and ommaya reservoir [9]. Shunt infections may be polymicrobial; the commonly encountered microbes are Gram positives such as Staphylococcus aureus, Staphylococcus epidermidis and Streptococcus pneumonia [10]. We have observed that infections of surgical hardware are surprisingly fewer in craniopharyngioma cases. Thus, to confirm this qualitative observation, we investigated using quantitative tests to elucidate the antimicrobial effects of cystic fluid from craniopharyngioma tumours.

Biological samples
Tumour cystic fluid was collected under sterile conditions on consecutive, histologically confirmed craniopharyngioma (adamantinomatous type) patients attending our clinic. Ethical approval was given and informed written consent was obtained from the parent/guardian for participation in the study (MRCZ/A/2608). Two samples were collected in their numerical order (1 and 2) over a period of 6 weeks and were stored at 4°C until testing. Patient demographics and relevant clinical history were also captured.

Preparation of test organisms
Three bacterial species common in our hospitals, namely, Escherichia coli (Gram negative), Staphylococcus aureus and Streptococcus pneumoniae (Gram positive), all clinical isolates, were selected as test organisms. E. coli and S. aureus were obtained from the Cimas Medical Aid Society laboratory and S. pneumoniae from Lancet laboratory. The test organisms were cultured in Mueller-Hinton broth overnight at 37°C. Immediately prior to the antibacterial assay, McFarland standard was used as a reference to adjust the inoculum size of each test organism to a concentration of 1 × 10 6 CFU/ml using sterile broth.

Bacterial growth inhibition assay
The antibacterial effect of the three samples against the selected human pathogens Escherichia coli, S. aureus and S. pneumoniae was determined using the agar disc diffusion method. Briefly, an aliquot of suspension containing 1 × 10 6 CFU/ml of bacteria was inoculated onto Mueller Hinton Agar (Mast Group Ltd., Merseyside, UK) and spread evenly on the agar surface. The sterile filter paper discs (Whatman No. 1, 6 mm) were soaked in 50 μl of each sample and then placed on the inoculated Mueller-Hinton agar plates. The discs were pressed gently to ensure complete contact with agar. Distilled water-loaded discs were used as negative controls whilst filter paper discs loaded with 20 μg/ml ampicillin or 50 μg/ml streptomycin were used as positive controls for the tested bacteria. The plates were stored at 4°C for 30 min to allow diffusion of extracts prior to incubation at 35 ± 2°C for 16 h. After incubation, the plates were observed for the formation of a clear zone around the discs which corresponds to the antibacterial activity of the tested extracts. The zones of inhibition (ZOI) observed were recorded as the diameter of the growth-free zones around the discs, measured in mm using a caliper. The antibacterial assay was carried out in triplicate.

Statistical analysis
Graph-pad Prism 6·01 was used to analyze the data. Frequency distribution of the numerical data was examined for normality and means, ± standard deviation was used as appropriate. Parametric data was analyzed using a one-way ANOVA across all groups. A p value ˂ 0.05 was considered to be statistically significant.

Results
Sample 1 was taken from a 9-year-old girl and sample 2 from an 11-year-old boy (Table 1). Both samples were from children of African descent diagnosed 1 and 3 years respectively prior to commencement of this study. Both patients had neurosurgical hardware insertion and displayed skin erosion over the ommaya shunt (see Figs. 1 and 2). On patient 1, the exposed part of the ommaya shunt was cultured and it grew Staphylococcus aureus; however, the intracystic part of the ommaya shunt, the cystic fluid and blood culture did not grow any pathogens, signifying contamination without infection. Both patients were infection negative. Neither of them showed any signs of infection on examination. Patient 2 reported an episode of the ommaya shunt falling-off and the mother, in ignorance, picked it up from the bathroom floor and stuck it back into the patient. Due to their rural abode, they reached the health care facility a week later. Both had not received antibiotics prior to presentation.
The effects of the craniopharyngioma cystic fluid collected from the two patients against the growth of E. coli, S. pneumoniae and S. aureus, expressed as the mean zone of inhibition diameters (mm), were notable (Table 2)

Discussion
This is the first study that demonstrates antimicrobial activity of the cystic fluid from craniopharyngioma pediatric patients of African descent. We report that the zones of inhibition by the craniopharyngioma cystic fluid of S. pneumoniae growth were higher than that of the commercial drug streptomycin.
These results confirm an inherent antimicrobial activity of the cystic fluid. Although ampicillin is known to inhibit Gramnegative bacteria like E. coli, the clinical bacterial isolate tested in this study showed resistance to the control antibiotic. S. pneumoniae was the most susceptible strain to the craniopharyngioma cystic fluid, followed by S. aureus whilst  There was notable absence of infection in both patients despite the skin erosion and increased risk for infection. On one patient, S. aureus was isolated and grown from the exposed part of the shunt, but no evidence of infection on cultures of the intracystic portion, the tumour fluid or blood cultures. The antibacterial activity against S. aureus of the craniopharyngioma cystic fluid could potentially explain this finding. These features have been reported to be associated   with infection in non-craniopharyngioma patients [11,12]. This was unexpected as we anticipated infection, especially in the patient that reported the shunt falling off. It is plausible that the cystic fluid has an inherent antibiotic activity. We propose that this antibiotic effect is coming from the alpha defensins which are components of the cystic fluid and have been shown to have antimicrobial properties [5,13]. Why the antimicrobial effect seems to be against Gram positives and not Gram negatives is not clear and would need further studies. Furthermore, possible explanations of the cystic fluid antimicrobial properties may be based on the biochemical makeup of the fluid. Previous reports on biochemical constituency of the craniopharyngioma cystic fluid have shown an increase in magnesium, alkaline phosphatase, blood urea nitrogen, glucose, urea and creatinine compared with serum levels in study participants [10]. Of note are magnesium and alkaline phosphatase. Magnesium was noted to be 40 times higher in craniopharyngioma cystic fluid and has been demonstrated to have in vitro antibacterial properties against E. coli, S. aureus and P. aeruginosa [14]. Alkaline phosphatase has been noted to have in vivo antimicrobial activity [15]. Thus, this study paves way for more investigations into the antimicrobial causative agents in the cystic fluid.
Whilst the sample size is small, considering the rarity of craniopharyngiomas and the infrequency of sampling of the cystic fluid, these cases, albeit few, presented a rare opportunity to interrogate further properties of the craniopharyngioma cystic fluid. This study may not allow bold statements to be made yet, but it undoubtably presents a unique finding. Another limitation may be that the storage of the cystic fluid albeit at 4°C may have unknown effects on cystic fluid composition.
The antibacterial activity of the tumour cystic fluid against E. coli, S. aureus and S. pneumoniae is of great importance as E. coli accounts for more than 70% of the infections of the urinary tract worldwide [16]. S. aureus is the most common cause of bacterial infections in abscesses of skin, joints and bones [17] whilst S. pneumoniae is the most common cause of community acquired pneumonia in children. Resistance to antibiotics has been reported for S. aureus, S. pneumoniae and E. coli [16,18]. Notably, the antibacterial properties of tumour cystic fluid could provide protection of host from shunt/hardware-related infections.

Conclusion
This study reports evidence of an antibacterial property of cystic fluid from craniopharyngioma patients. This is an important property in preventing infections in these patients and may also help to select appropriate empirical antibiotics when infection occurs, which is of paramount importance. Further biochemical investigations into the properties of cystic fluid that makes it have antibacterial properties are recommended.
Acknowledgements Cimas Medical Aid Society and Lancet laboratories for providing the bacterial strains used in this study.
Authors' contributions LJ, TR, TLMJ and TM conceived and designed the study. TLMJ, TR, MK and TM performed the clinical examination or microbiology and the data analysis. LJ wrote the first draft and all authors read, edited and approved the final manuscript.

Funding information
The study received material support from the Biochemistry Department, University of Zimbabwe.
Data availability The statistical data on the microbiology and clinical scores used to support the findings of this study are available from the corresponding author upon request.

Compliance with ethical standards
Conflict of interest The authors declare that there is no conflict of interest.
Ethics approval and consent to participate Ethical approval for the study was obtained from Medical Research Council of Zimbabwe (MRCZ/A/2608). Gatekeeper approval was obtained from Bioethics Committee of Harare Central Hospital, Zimbabwe. Written informed consent was obtained from the parents or guardians of the children. Written assent was obtained from children.