This was an unmatched cross sectional case control study. Approval for the study was obtained from the University of Zambia Biomedical Research Ethics Committee (006-03-19, 25th April, 2019).
Participants, both PD and non-PD were black Zambians 18 years of age and above. PD participants who met the UK Parkinson’s Disease Society Brain Bank clinical diagnostic criteria were recruited, regardless of duration of the disease (9). Controls were recruited from members of the patient’s environment including spouses of PD patients who were free of neurodegenerative disease, healthy adults looking after in hospital patients, and healthy medical staff.
Written, informed consent was obtained from participants. From the study were excluded, individuals that declined consent, had parkinsonism, including history of undergoing treatment with medication that might induce parkinsonism (e.g. metoclopramide, typical and atypical antipsychotic agents), had stroke-related parkinsonism, head injury or sub-arachnoid hemorrhage, history of encephalitis, evidence of chronic liver disease, GI malignancy, diagnosis of inflammatory bowel disease and a history of GI motility drugs use.
Adults attending the neurology clinic at the University Teaching Hospital (UTH), Adult Hospital, and those admitted via the Adult Medical Emergency Unit (AMEU) and medical wards were invited to participate and recruited consecutively.
The University Teaching Hospital, Adult Hospital in Lusaka is a third level hospital and functions as the country’s national referral center as well as a provincial and district hospital.
The background assumption for LBP was 26.2ng/ml based on local studies and we assumed a 20% increase in adults with PD (31.4ng/ml), with standard deviations of 6.9 in both groups (10). We had initially assumed a 90% power at 95% confidence to detect a 20% increase in LBP in PD cases and that gave us a sample size of 38 cases for each group. However with the onset of the pandemic of covid 19, we encountered challenges meeting the target sample size as clinics were closed in order to safeguard the welfare of the participants. We thus assumed a ratio 1:2, so that for each case frequency there were two unmatched controls instead of the prior 1:1 and assumed a power of 80% with and alpha of 0.05. The resulting sample size was 22 cases with 44 controls. Stata was used for the calculations.
During recruitment a thorough clinical assessment was carried out in which, patient’s demographics including age, sex, predominant PD phenotype, residency, occupation, use of tap water, use of medication for PD, family history of PD, and other comorbidities were documented using a questionnaire. Physical and neurologic examination was carried out by the investigator.
Following the clinical assessment and examination, about 5mls of blood was drawn from each enrolled participant via venipuncture and placed into EDTA bottles. The blood was centrifuged at 2700rpm for 15 minutes to extract plasma, within 24hrs and the plasma was stored at ¯80°c in the Department of Internal Medicine laboratory (TROPGAN freezer). Samples were thawed and processed using the DuoSet ELISA Human LBP Elisa kit as per standard procedure. Samples were run in duplicate concurrently with standards.
DNA Extraction Method Using Qiagen Kit
Plasma samples were equilibrated at room temperature (15–25°C) after which 20µl of QIAGEN protease enzyme was added to each 200µl of plasma sample in 1.5ml microfuge tubes. Samples which were less than 200µl were adjusted by adding appropriate volume of PBS and all samples were properly mixed after adding the enzyme. 200µl of buffer AL (containing 28µg/ml of carrier RNA) was added followed by vortexing for 15 seconds to ensure efficient lysis. The samples were incubated at 56°C for 10 minutes after which they were centrifuged to remove drops from inside the lid followed by addition of 250µl of absolute ethanol and thorough mixing for 15 seconds. The mixtures were transferred to the 2 ml QIAamp spin column and centrifuged at 8000 rpm for 1 minute, and the filtrate was discarded. The QIAamp spin columns were transferred to clean collection tubes, and each spin column was closed to avoid aerosol formation during centrifugation. Centrifugation did the cleaning of the samples was at higher speed until the QIAamp spin column was empty. Then 500 µl buffer AW1 were carefully added followed by centrifugation at 8,000 rpm for 1 minute. Spin columns were after that transferred to clean 2 ml collection tubes and 500 µl buffer AW2 was added without wetting the rim. The caps were closed, and the columns were again centrifuged at 8,000 rpm for 1 minute. Spin columns were moved to a clean 2 ml collection tubes and, 500 µl of absolute alcohol was added followed by centrifugation at 8,000 rpm for 1 minute. QIAamp columns were transferred to clean 2 ml collection tubes and centrifuged at 14,000 rpm to dry them. The columns were transferred to a clean 1.5 ml microcentrifuge tube and 200 µl buffer AVE was added and they were incubated at 56ºC for 5 minutes to increase DNA yield. DNA was eluted by centrifuging the columns at 14,000 rpm for 1 minute. The concentration and purity of DNA was spectrophotometrically determined using Nanodrop spectrophotometer 2000c (Thermo scientific, Massachusetts, USA). DNA was kept at -20ºC until needed.
Recovery efficiency of extraction method
The efficiency of the DNA extraction method was assessed by using, a known clone, 20µl of DNA from K12 American Type Culture Collection (ATCC) 92 522 with the concentration of 6.37ng/µl was added to 180µl of plasma sample from a donor considered healthy (with no bacteria in blood) followed by DNA extraction of DNA as described above. After extraction of DNA, the concentration was measured (5.3ng/µl) and the purity (260/280) was determined as 1.73 using Nanodrop spectrophotometer 2000c. The experiment was repeated three times, and the average efficiency was 5.3/6.37 or 83.2%.
Real-time quantitative PCR (q-PCR)
Real-time quantitative PCR was carried out using broad range primers for 16S rRNA. All primers were reconstituted with sterile water to final stock concentrations of 1mg/ml while the working primer concentrations were 50 nano grams. Real time quantitative polymerase chain reaction mixes were prepared quantitative PCR detection was performed using QuantiTect SYBR® Green PCR kit from Qiagen following the manufacturer's instructions. The qPCR was performed on a Rotor gene 6000 real time quantitative PCR machine (Rotor gene, North West, Australia). The reaction temperature profile consisted of an initial denaturing at 95C for 15 minutes, cycle denaturation at 95C for 15 seconds, annealing was at 60C for 30 seconds and extension at 72C for 45 seconds.
Real-time PCR
Real-time quantitative PCR was carried out using broad range primers for 16S rRNA. The Primers were reconstituted with sterile water, and the forward primer 16S F519 (5'-CAGCAGCCGCGGTAATAC-3') with the total of 249.8µg, 249.8µl of sterile water was added making a stock concentration of 1mg/ml. The reverse primer 16S R785 (5'-TGGACTACCAGGGTATCTAATCC-3') with the total of 345.2µg, 345.2µl of sterile water was added making a stock concentration of 1mg/ml. The primers were both diluted to 50ng/ml (20-fold) to get working concentration. The addition of 10µl of each primer (forward and reverse) from the stock was added to 180µl of sterile water to make 200ul.
Determination of specificity and sensitivity of the 16S real-time quantitative PCR assay
Determination of the specificity of the 16S rRNA gene primers was done using a known quantified (E.coli K12 strain). A 10-fold serial dilution of a quantified plasmid was used to establish the standard curve which was used for absolute quantification starting with 108, 107, 106, 105, 104 and 103 16S rRNA copy number per microliter (µl) and Non-Template Control (with each dilution tested in triplicate). After PCR amplification had been done, the sensitivity of PCR and standard curve was generated. The melting curves were generated and showed sharp peaks at the expected melting temperature (85.4°C ± 1.3) indicating specific detection 16s rRNA gene product.
Melting curve analysis for confirmation of amplicons
Melting curve analysis temperature of the PCR was increased, and as it was raised over a gradient, the double stranded template was denatured resulting in the SYBR green dissociate from the product and matching fall in fluorescence. At every point fluorescence and temperature intensities were measured and confirmed that targeted template was present, 50% of the double stranded template at a temperature of 83.4ºC was denatured. A melting curve analysis was carried out over a period of 20 minutes by first cooling the reaction to 60°C then heating to 99°C. Melting curve for each sample was constantly quantified during the heating process and the SYBR Green I fluorescence was plotted against temperature. The peak point in the melting curve analysis was deemed to be the melting temperature (Tm) after plotting the negative offshoot of the fluorescence aligned with temperature (-dF/dF vs. T).
Data Analysis
Data were entered on designed data collection tool then double entered onto Microsoft Office excel spread sheet using patient identification number analyzed using STATA 15.
Descriptive statistics of means and standard deviation, or frequency distribution and percentages was used to present demographic and clinical characteristics data that were continuous or categorical respectively. Where data were skewed, medians with interquartile ranges was used.
Categorical data from PD patients and controls were analyzed for association using Fisher’s exact test due to the small cell values. The 16S rRNA was dichotomized and Fisher’s exact test used to analyze for association between PD cases and controls. Continuous data for both PD cases and controls was analyzed for statistically significant differences with the Student’s t-test when data were parametric and Mann-Whitney U test in non-normally distributed data. A level of significance less than 5%, (p < 0.05) was applied throughout.