Drugs and chemicals
Streptozotocin was purchased from Sigma Aldrich (St. Louis, MO, USA). All other chemicals used were of analytical grade.
Plant collection and preparation of plant extract
Dried roots of C. borivilianum were procured from Nandan Agro Farms Pvt. Ltd. (Hyderabad, Andhra Pradesh, India) and authenticated by senior botanist, Madhava Setty, Sri Venkateswara University, Tirupati, India. The plant was deposited at Herbarium with the number: KLU 96568. The dried roots were cut into small pieces and grounded into fine powder. The root powder (1000 g) was subjected to cold maceration in 2 L of sterile distilled water for 48 hours at room temperature, filtered into a clean round bottom flask using adsorbent cotton wool and a filter paper (Whatman No. A-1). The whole process was repeated seven times to ensure maximum yield of water soluble compounds from the root powder. The combined aqueous extract was concentrated at 37°C using a rotary evaporator (R-210, Buchi Labortechnik AG, Flawil, Switzerland) and lyophilized by using a cyodos freeze dryer (Telstar, Barcelona, Spain) to yield approximately 38 g of solid extract (3.8% w/w).
A qualitative phytochemical evaluation was performed on the aqueous root extracts to determine the presence of carbohydrates (Barfoed’s test), flavonoids (test of Shinoda), phytosterols (Libermann Buchard test), phenols (ferric chloride test), alkaloids (Dragendorff test), proteins (Biuret test) and saponins (Saponification test) following the methods as described by Harbourne .
Fourier transform infrared (FTIR) spectroscopy
FTIR spectroscopic analysis was performed using Perkin Elmer spectrophotometer system (PerkinElmer, Inc., Shelton, CT, USA) which detects the characteristic peaks and functional groups that are present in the root extract. The spectral range 4000 to 600 cm-1 with resolution of 2 cm-1 was used to record the infrared spectra with the potassium bromide (KBr) pellet making technique.
Adult male Wistar rats (170–200 g) were obtained from Animal House, Faculty of Medicine, University of Malaya (UM). The rats were kept under standard environmental conditions of room temperature 25 ± 2°C, relative humidity between 45-55% and 12 hrs light/dark cycle. Animals were fed with standard feed pellets (Harlan diet, UK) and tap water ad libitum. Experimental procedures were in accordance with ARRIVE guidelines (Animals in Research: Reporting In-Vivo Experiments) and European Community Guidelines/ EEC Directive, 1986. This study was approved by the Faculty of Medicine, Animal Care and Use Committee (ACUC), University of Malaya with ethics number: 2013-07-15/FIS/R/NS. Toxicity study was conducted according to Organization for Economic Cooperation and Development (OECD) revised up-and-down procedure for acute toxicity testing (OECD guideline 425) . No signs of toxicity were observed in the tested animals up to a dose of 3000 mg/kg/day.
Induction of diabetes
Hyperglycemia was induced in overnight fasted male rats by a single intraperitoneal injection of STZ dissolved in ice cold citrate buffer (0.1 M, pH 4.5) at a dose of 55 mg/kg . The rats were allowed to drink 5% sucrose solution overnight after injection, to overcome drug-induced hypoglycemia. Diabetes was confirmed by the presence of polydipsia, polyuria and weight loss and only animals exhibiting a fasting glucose level greater than 300 mg/dL three days after STZ injection were used in this study. Treatment with C. borivilianum was commenced four days after STZ injection and this was considered as day one. C. borivilianum root extract at 250 and 500 mg/kg/day  were administered in the form of suspension in 1% sodium carboxymethylcellulose (Na-CMC) dissolved in distilled water daily for 28 consecutive days using an oral gavage tube.
The animals were randomly assigned into five experimental groups with six (6) rats per group:
Group I - Control rats- received 1% Na-CMC vehicle only.
Group II - Diabetic control rats- received 1% Na-CMC vehicle only.
Group III and IV - Diabetic rats treated with C. borivilianum root aqueous extract at 250 & 500 mg/kg body weight respectively.
Group V- Diabetic rats treated with standard antidiabetic agent, glibenclamide at 600 μg/kg body weight.
At the end of experimental period, all rats were fasted overnight, weighted and sacrificed under pentobarbital sodium anesthesia (60 mg/kg) followed by cervical dislocation.
Determination of sperm characteristics and morphology
Immediately after euthanasia, cauda epididymis were dissected out, chopped and placed in 5 ml physiological saline (0.9% NaCl) and incubated for 5 min at 37°C in water bath to allow sperm to leave the epididymal tubules.
Evaluation of sperm forward motility
Progressive sperm motility was evaluated by a method of Belsey et al. . Firstly, immotile sperm were counted followed by motile sperm. Sperm forward motility was expressed as percentage of motile sperm to total sperm counted.
Sperm count was determined using Neubauer chamber (Deep 1/10 mm, LAMBART, Darmstadt, Germany) of hemocytometer following the method as described by Belsey et al. . Sperm count was expressed as number of sperm per ml of solution.
The ratio of live to dead sperm was determined using 1% trypan blue staining following the method as described by Talbot and Chacon . A total number of 200 sperm were counted per slide and the results were expressed as percentage of the live sperm.
Hypo-osmotic Swelling Test (HOST)
Sperm’s flagella membrane integrity was assessed by hypo-osmotic swelling test (HOST). In brief, assay was performed by incubating 50 μL sperm suspension with 1 mL hypo-osmotic solution. Two hundred sperm were evaluated and percentage of live sperm (with coiled tail) was calculated following the method of Jeyendran et al. .
Sperm morphological abnormalities
Percentages of sperm head, middle piece and tail abnormalities were determined from a total of 300 sperm per rat . Sperm morphology was viewed under a light microscope (Nikon, H600L, Tokyo, Japan) under 400 × magnifications. Data was expressed as percentage of morphologically abnormal sperm to total sperm count.
Following analysis of sperm parameters, the remaining cauda epididymal sperm suspension was centrifuged at 800 g for 20 min at 4°C and the pellet was resuspended in normal saline. The sperm were homogenized with a glass-Teflon Homogenizer (Heidolph Silent Crusher M, Germany). The supernatant collected was used for biochemical studies.
Assessment of sperm lipid peroxidation
The lipid peroxidation level in sperm homogenate was measured as malondialdehyde (MDA), which is the end product of lipid peroxidation, which reacts with thiobarbituric acid (TBA) to produce TBA reactive substance (TBARS), a red colored complex which has peak absorbance at 532 nm, according to the method of Ohkawa et al., .
Determination of Hydrogen peroxide and nitric oxide level in sperm
In the supernatant levels of H2O2 were estimated according to the method by Pick . In brief, H2O2 mediates horseradish peroxidase-dependent oxidation of phenol red and the results obtained were expressed as nmol H2O2 formed g-1 tissue. Nitric oxide (NO) concentration was measured as nitrite/nitrate by previous method of Miranda et al., . The NO levels were expressed as nmol/g tissue.
Assessment of Sperm Total Antioxidant Capacity (TAC)
The total antioxidant capacity of sperm homogenate was measured according to the modified method of Erel, , where the samples were mixed with acetate buffer, and 2,2’-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS reagent) with hydrogen peroxide. The color reaction was monitored spectrophotometrically at 660 nm. The results were expressed as μmolequivalents Trolox/mg protein.
Determination of sperm endogenous antioxidant enzyme activity
SOD activity level was assayed according to the method by Misra and Fridovich  and was expressed as the amount of enzyme that inhibits oxidation of epinephrine by 50%, which was equal to 1U per milligram of protein. CAT activity level was determined on the basis of H2O2 decomposition  and was expressed in μmol of H2O2 metabolized/mg protein/min. Meanwhile, GPx activity level was determined according to the method by Rotruck et al.  and was expressed as μmol of GSH consumed/mg protein/min.
In-vitro radical scavenging activity of C. borivilianum root extract
Antioxidant power of C. borivilianum root aqueous extract was evaluated by using DPPH, superoxide, hydroxyl, H2O2, NO and reducing power scavenging assays. Ascorbic acid (10–200 μg/ml) was used as a standard. The ability of extract to scavenge or inhibit free radicals was expressed as % inhibition and was calculated using this formula.
Ao = absorbance of control group (without plant extract) and At = absorbance of C. borivilianum root aqueous extract. Absorbance was measured using a spectrophotometer (UV-1700, Shimadzu, Kyoto, Japan) at different wavelengths depending on the type of assay performed. All assays were carried out in triplicate.
Radical-scavenging activity of C. borivilianum root aqueous extract against stable DPPH radical was determined according to the method by Katalinic et al., . Measurement of superoxide radical scavenging activity was based on the method by Xiang and Ning . Hydroxyl radical scavenging activity was measured using a modified method by Halliwell et al., . The ability of C. borivilianum root aqueous extract to scavenge H2O2 was determined according to the method by Ruch et al., . The ability of the extract to scavenge nitric oxide was determined using a method by Dastmalchi et al., . Finally, reducing power scavenging activity of C. borivilianum root extract was determined according to the method by Suseela et al., .
Estimation of sperm caspase-3 levels
Snapped-frozen sperm were homogenized using a sonicator with PRO-PREP (iNtRON Biotechnology, Seoul, South Korea) extraction solution in the presence of protease inhibitors. Total cell protein was obtained by centrifugation at 13000 g for 15 minutes at 4°C. After determination of the protein concentration, the same amount of protein was loaded into the 12% SDS-PAGE gel. The protein was then transferred onto the polyvinyilidene difluoride (PVDF) membrane and incubated in 5% BSA for 90 minutes. The blot was then exposed to primary antibody, caspase-3 rabbit polyclonal IgG, Santa Cruz, USA (sc-7148) at 1:1000 dilution. Following primary antibody incubation, the blot was incubated with HRP-conjugated secondary antibody and finally visualized by using Optic 4C (Bio RAD). β-actin (Santa cruz, sc-130656) was used as a loading control. Photos of the blots were captured and density of each band was determined using Image J software (1.39, Bethesda, Maryland, USA). The ratio of caspase-3/ β-actin band density was determined and was considered the expression level of each of the target proteins.
Determination of Serum HbA1c and Fasting Blood Glucose (FBG) Levels
At the end of 28 day treatment, blood was withdrawn from retro-orbital plexus in overnight fasted diabetic rats. Glucose levels were estimated using glucose oxidase/peroxidase kit (BioSystems S.A. Costa Brava 30, Barcelona, Spain) and HbA1c was measured using a commercially available kit (BioSystems S.A. Costa Brava 30, Barcelona, Spain). The values were estimated using an automated analyzer Dimension RxL Max Integrated Chemistry System (Siemens Healthcare Diagnostics Inc. Deerfield, IL, USA).
Epididymal sperm density
Caput epididymis was freshly harvested and fixed in 10% formaldehyde for 24 hours. The tissue was dehydrated in ethanol, cleared in xylene and embedded in paraffin at 58°C. The embedded tissue samples were cut into 5 μm thickness using a microtome (Histo-line laboratories, ARM-3600, Viabrembo, Milano, Italy). After deparaffinization by immersion into xylene for 20 min, tissues were dropped in ethanol solution of decreasing concentrations (100%, 95%, 90% and 80%) for 5 min. Sections were then stained with hematoxylin and eosin (H&E). Sperm density in epididymal lumen was graded as normal (+++), moderately decreased (++), or severely decreased (+) according to description by Narayana et al., .
Statistical differences were evaluated by analysis of variance (ANOVA) and Student’s t-test. A probability level of less than 0.05 (p < 0.05) was considered as significant. Post-hoc statistical power analysis was performed for all the experiments and values > 0.8 were considered as adequate. Meanwhile, Shapiro-Wilk test was performed to test for data normality and values > 0.05 indicated that data were normally distributed.