In vitro pharmacological interaction of caffeine and first-line antibiotics is antagonistic against clinically important bacterial pathogens

The in vitro antibacterial activity of pure caffeine powder and its interaction with first line antibiotic against bacterial isolates were investigated with the macrobroth dilution and the checkerboard assay methods. This study showed that caffeine and the antibiotics exhibited various degrees of antibacterial activities. While caffeine had MICs ranging between 67.19 and 268.75 μg/ml, chloramphenicol was characterized by MICs between 0.98 and 31.25 μg/ml, kanamycin — 15.63–62.5 μg/ml, nalidixic acid — 0.49–250 μg/ml, erythromycin — 0.49–62.5 μg/ ml, tetracycline — 1.99–62.5 μg/ml and metronidazole — 15.63–31.25 μg/ml. Combining 1⁄2 MICs and MICs of caffeine with the antibiotics as well as direct combination of caffeine and the antibiotics resulted in significant reduction of antibiotics’ effectiveness. The fractional inhibitory concentration index (FICI) for the combination of 1⁄2 MICs of caffeine with different antibiotics showed antagonistic interactions with the antibiotics except kanamycin which had additive and indifferent interactions with caffeine. The FICI of the MICs of caffeine combined with antibiotics showed a reduction in the number of antagonistic interactions as chloramphenicol, nalidixic acid and erythromycin showed some indifferent interactions while kanamycin was the only antibiotic that showed indifferent interaction against all the bacterial isolates. The direct combination of caffeine and the antibiotics resulted in significant antagonistic interactions higher than in the case when caffeine, at the 1⁄2 MICs and MICs, was combined with the antibiotics. Although caffeine demonstrated significant antibacterial activity against the selected bacterial isolates, its combination with the selected antibiotics resulted in significant antagonistic interactions. Caffeine should not be combined with antibiotics as this could result in serious therapeutic failure and, possibly, drug toxicity in vivo.


INTRODUCTION
, is a white powdered, water soluble plant alkaloid found in many plant species.This psychoactive, widely consumed as a stimulant (Belitz et al., 2009;Mohanpuria et al., 2010), is a purine alkaloid present in almost 100 plant species (Ashihara, 2006).Natural sources of caffeine include cocoa (DeVries et al., 1981), coffee (Minamisawa et al., 2004) and tea (Najafi et al., 2004).The caffeine content in fresh coffee ranged between 314 and 646 mg/l (Rodrigues et al., 2007) and the content for brewed tea including black tea (30.97 mg/g), green tea (18.70 mg/g) and oolong tea (23.89 mg/g) was reported by Guo and coworkers (2011), whereas caffeine content in cocoa powder was between 0.66 and 0.71 mg/g (Li et al., 2012).Caffeine, a white odorless powder with a bitter taste, has a molecular weight of 194.191 g/ mol, a density of 1.2 g/cm 3 and pH of 6.9.The solubility of caffeine in water is about 21.7 mg/ml at 25°C, 180 mg/ml at 80°C and 670 mg/ml at 100°C.Its boiling point is at 178°C and its melting point is at 238°C (Florey et al., 1989;Agyemang-Yeboah et al., 2013).
Caffeine has long been known to have numerous effects on human health (Cano-Marquina et al., 2013).In humans, it is mainly metabolized into paraxanthine (84%), theobromine (12%) and theophylline (4%) (Lelo et al., 1986;Etherton & Kochar, 1993).The daily average intake of caffeine through coffee consumption in the US is between 165 and 200 mg/day (Mitchell et al., 2014), with adults taking about 2.4 mg per kg per day and children aged between 5 and 18 years old taking 1.1 mg per kg per day (Chou et al., 1992;Desbrow et al., 2007), while its daily intake in about four to nine cups of coffee is equivalent to 500 to 600 mg.This is recognized as an abuse posing serious health risks known as caffeinism which could result in caffeine dependency, headaches, restlessness, heart palpitations, nervousness, insomnia, vomiting, nausea and diarrhea (James et al., 1985), stimulation of the central nervous system (Nehlig et al., 1992) as well as the release of the stress hormones.When these hormones are chronically elevated, mental concentration is short-lived and fine motor coordination is impaired (Jacobson & Thurman-Lacey, 1992), the immune system is suppressed, digestion and elimination are impaired and body's normal repair mechanisms are inhibited, thereby accelerating aging process (Hill, 1991;Raber, 1998;Leonard, 2000).When stimulatory effects of caffeine are reduced significantly, there is tolerance adaptation.Due to adaptive reactions to caffeine, individuals become more sensitized to adenosine.A reduction in caffeine consumption effectively increases the normal functional effects of adenosine.This could result in undesirable withdrawal indications in tolerant users.Withdrawal symptoms may appear within 12 to 24 h after termination of caffeine intake.They may last for a maximum of five days, which is the sufficient time, required for the adenosine receptors in the brain to revert to normal levels while mental awareness and general body coordination are restored.
Pharmacologically, caffeine inhibits muscle contraction (Youn et al., 1991), acts as the antagonist of the adenosine receptors to increase their plasmatic concentration (Conlay et al., 1997), potentiates the lethal effects of ionizing radiation which is important during the treatment of cancer (Sakurai et al., 1999), alters glucose metabolism (Greer et al., 2001), sensitizes calcium liberation channels and inhibits phosphodiesterase enzymes (Daly, 2007).Pure caffeine has antibacterial effect (Ramanaviciene et al., 2003a;Cogo et al., 2008;Mohammed & Al-Bayati, 2009), decreases the adherence of Streptococcus mutants to dental surface (Namboodiripad et al., 2007) and acts as the first dietary component able to protect one against Alzheimer's disease, Parkinson's disease and Huntington's disease (Ross et al., 2000;Lindsay et al., 2002;Ribeiro & Sebastiao, 2010).It increases the concentration of some immunocompetent cells and reinforces the activity of lysozyme to boost immunity against bacterial pathogens (Ramanaviciene et al., 2002;Ramanaviciene et al., 2003b;Vinod & Rangari, 2004).Although there are interactions between A2A receptors and the dopaminergic system in the brain, as adenosine hinders dopaminergic neurotransmission, the inhibition of A2A receptors by caffeine may accelerate dopaminergic activity and worsen psychotic symptoms (Kruger et al., 1996) and as one of the 10 most frequently administered drugs in neonatal intensive care, caffeine is used to correct irregular heartbeat and treat apnea in premature newborns (Funk et al., 2009).
There are 82 known drug interactions with caffeine (Prescription drugs/caffeine interactions, 2016.http:// www.caffeineinformer.com/caffeine-drug-interactions.Accessed 01-05-2016).Caffeine increases the activity of some pain killers such as acetaminophen, tylenol and aspirin.It enhances their effectiveness and brings quicker relief by allowing rapid absorption of these drugs into the body.As a result, many over-the-counter headache drugs contain caffeine in their formula.Its combination with ergotamine in the treatment of migraine and headache clusters helps overcome the drowsiness caused by antihistamines (Nehlig et al., 2000).While caffeine works synergistically with some pain relievers (Hidron et al., 2007), it hinders the proper functioning of other drugs like fluvoxamine, which impedes the action of liver enzyme responsible for the breakdown of caffeine and, thus, increases the central effects and blood concentration of caffeine 5 times (Hidron et al., 2007).Combining caffeine and ephedrine, which are both stimulant drugs, could result in side effects and heart problems.Taking antibiotics along with caffeine can increase the risk of having side effects (Coso et al., 2012), fluoroquinolones decrease the rate of caffeine breakdown by the organism (Hidron et al., 2007).As there is the dearth of information on the in vitro antibacterial action of pure caffeine powder and its interaction with first line antibiotics, this study investigated the effects of pure caffeine in the form of powder on the antibacterial activity of some first line antibiotics against bacterial isolates.Preparation of the samples.The antibiotics were prepared according to manufacturers' specification.Stock solutions of each of the drugs: kanamycin, tetracycline, chloramphenicol, nalidixic acid, erythromycin and caffeine, were prepared by dissolving each drug in its respective solvent.

Test drugs.
Determination of the minimum inhibitory concentrations (MICs).The minimum inhibitory concentrations (MICs) of the caffeine and those of each of the antibiotics were determined by macrobroth dilution bioassay (Olajuyigbe & Afolayan, 2012).Overnight cultures were diluted with sterile nutrient broth.To determine the MICs of each of the drugs alone, one milliliter of stock solution of caffeine was serially diluted in sterile double strength Mueller Hinton broth in test tubes to obtain concentrations ranging between 1.05 and 537.5 µg/ml while concentrations ranging between 0.488 and 250 µg/ ml were prepared for all the antibiotics.To determine the influence of the minimum inhibitory concentrations of caffeine on the antibacterial activities of the antibiotics, different concentrations of each of the antibiotics were serially prepared in double strength Mueller Hinton broth while the minimum inhibitory concentration (MICs) and half MICs ( 1 / 2 MICs) of caffeine were added to respective tubes containing different concentrations of each antibiotic.For the combination of caffeine and the antibiotics, equal volumes of stock solutions of caffeine and each antibiotic were combined before being serially diluted to concentrations corresponding to the MICs of each drug in double strength Mueller Hinton broth.After the drug preparations were serially diluted, 100 µl of each diluted bacterial culture was aseptically dispensed to the test tubes containing caffeine alone, antibiotics alone and their combinations and incubated at 37°C for 24 h.Bacterial growth was indicated by the turbidity of the test tube.The MIC values were recorded as the lowest concentration of each antibiotic showing no visible growth.
Checkerboard assay.The interactions between caffeine and antibiotics were determined using the checkerboard assay.The drug concentrations used in the checkerboard assay covered the MIC for each drug used in the assay.The fractional inhibitory concentration (FIC) was derived from the lowest concentration of the caffeine and the antibiotic used in combination and resulting in no visible growth of the test organisms in the Mueller-Hinton broth after incubation for 24 h at 37°C (Mandal et al., 2004).FIC indices (FICI) were calculated using the formula: FIC index = (MIC of caffeine in combination/ MIC of caffeine alone)+(MIC of antibiotic in combination/MIC of antibiotic alone).In antimicrobial combination, Eliopoulos and Eliopoulos and coworkers )1988) and Petersen et al., (2006) defined synergy as ΣFIC ≤ 0.5, additivity as 0.5 < ΣFIC ≤1, indifference as 1 < ΣFIC ≤ 4 and antagonism as ΣFIC > 4.This implies that synergy In vitro antibacterial effects of caffeine combined with antibiotics determined by the checkerboard method may be defined as FIC ≤ 0.5 or FIC ≤ 1.

RESULTS
In this study, we determined the antibacterial activity of caffeine and first line antibiotics as well as the influence of caffeine on the antibacterial activity of the antibiotics.Table 1 shows the minimum inhibitory concentrations for caffeine and antibiotics used alone against the selected bacterial isolates.The caffeine and the antibiotics exhibited various degree of antibacterial activity.Caffeine had MICs ranging between 67.19 and 268.75 µg/ml, chloramphenicol had MICs of between 0.98 and 31.25 µg/ml, kanamycin -15.63-62.5 µg/ml, nalidixic acid -0.49-250 µg/ml, erythromycin -0.49-62.5 µg/ ml, tetracycline -1.99-62.5 µg/ml and metronidazole -15.63-31.25  Table 2 shows the influence of caffeine at half of the minimum inhibitory concentrations ( 1 / 2 MICs) on the antibacterial activity of the antibiotics.Combining caffeine with the antibiotics at its half MICs resulted in significant increase of the MICs of the antibiotics.Chloramphenicol that inhibited the bacterial isolates at concentrations ranging between 0.98 and 125 µg/ml when used alone had MICs ranging between 3.91 and 250 µg/ ml when used in combination with caffeine at its half MICs.At 31.25 µg/ml, this antibiotic inhibited the growth of P. vulgaris CSIR 0030 and E. coli ATCC 8739.However, Ps. aeruginosa ATCC 19582 and B. cereus ATCC 10702 were inhibited at 3.91 µg/ml, S. sonnei ATCC 29930, K. pneumonia ATCC 10031, Ps. aeruginosa ATCC 15442 and E. faecalis ATCC 29212 at 250 µg/ml while P. shigellosis ATCC 379003 and E. cloacae were inhibited at 125 µg/ml.
Kanamycin having MICs ranging between 15.63 and 62.5 µg/ml when used alone had its MICs ranging between 1.95 and 62.5 µg/ml when combined with caffeine at its ½ MICs.It inhibited the growth of P. vulgaris CSIR 0030 and E. coli ATCC 8739 at 1.95 µg/ml, K. pneumoniae ATCC 10031 and E. faecalis ATCC 29212 at 15.63 µg/ml, P. shigellosis ATCC 379003 and E. cloacae ATCC 13047 were inhibited at 31.25 µg/ml while Ps.aeruginosa ATCC 15442 and Ps.aeruginosa ATCC 19582 were inhibited at 62.5 µg/ml.While nalidixic acid used alone had MICs ranging between 0.49 and 250 µg/ ml, its combination with caffeine ( 1 / 2 MICs) resulted in its inhibitory concentrations, effective against all the test bacterial isolates, being greater than 250 µg/ml.On the other hand, erythromycin had MICs of 0.49 to 62.5 µg/ml when used alone with most of the isolates being inhibited at 0.49 µg/ml and it had its inhibitory     62.5 and 250 µg/ml when combined with caffeine at its half MICs.Upon the combination of metronidazole with caffeine at ½ MICs, E. faecalis ATCC 29212, E. cloacae ATCC 13047 and B. cereus ATCC 10702 were inhibited at 62.5 µg/ml and E. coli ATCC 8739 was inhibited at 250 µg/ml while other bacterial isolates were inhibited at 125 µg/ml.Table 3 shows the influence of the MIC of the caffeine on the antibacterial activity of the antibiotics.In combinations with caffeine, chloramphenicol had MICs ranging between 15.63 and 125 µg/ml.It inhibited P. vulgaris CSIR 0030 at 15.63 µg/ml, E. cloacae, Ps. aeruginosa ATCC 19582 and P. shigellosis ATCC 379003 at 31.25 µg/ml, E. coli, B. cereus ATCC 10702 and Ps.aeruginosa ATCC 15442 at 62.5 µg/ml, and S. sonnei ATCC 29930, E. faecalis ATCC 29212, K. pneumoniae ATCC 10031 at 125 µg/ml.While the interaction between caffeine and kanamycin resulted in MICs ranging between 0.98 and 62.5 µg/ml for kanamycin, it inhibited Ps. aeruginosa ATCC 19582 at 0.98 µg/ml, Proteus vulgaris CSIR 0030 at 1.95 µg/ml, E. cloacae at 7.81 µg/ ml, E. coli ATCC 8739, Ps. aeruginosa ATCC 1542, S. sonnei ATCC 29930 and B. cereus ATCC 10702 at 31.25 µg/ ml, whereas E. faecalis ATCC 29212 and K. pneumoniae ATCC 10031 at 62.5 µg/ml.For nalidixic acid, its MICs ranged between 125 and > 250 µg/ml when combined with caffeine at its MICs.With the exception of P. shigellosis ATCC 379003 being inhibited at 125 µg/ml and E. cloacae ATCC 13047, E. coli ATCC 8739 and P. vulgaris CSIR 0030 being inhibited at concentrations greater than 250 µg/ml, other isolates were inhibited at 250 µg/ ml when nalidixic acid was combined with MICs of caffeine.Combining erythromycin with caffeine increased the MICs up to 7.81-> 250 µg/ml with most of the isolates being inhibited at 250 and > 250 µg/ml.While the combination of caffeine at MICs with tetracycline resulted in tetracycline's MICs being increased to between 7.81 and 250 µg/ml, those of metronidazole was between 125 and > 250 µg/ml with most of the isolates being inhibited at 250 and > 250 µg/ml except E. cloacae ATCC 13047 being inhibited at 125 µg/ml.The direct combination of caffeine with the antibiotics resulted in significant increase in the minimum inhibitory concentrations of the antibiotics against all the bacterial isolates as shown in Table 4.
In Tables 5 to 7, the fractional inhibitory concentration index was used to find out if the interactions between caffeine and the antibiotics were antagonistic, synergistic, additive or indifferent.The fractional inhibitory concentration of the combination of ½ MICs of caffeine with the different antibiotics showed that all the antibiotics exhibited antagonistic except kanamycin having additive and indifferent interactions with caffeine as shown in Table 5.However, combining the antibiotics with the MICs of caffeine showed a reduction in number of antagonistic interactions as chloramphenicol, nalidixic acid and erythromycin showed some indifferent while kanamycin was the only antibiotic that showed indifferent interaction against all the selected bacterial strains as shown in Table 6.The direct combination of caffeine and each of the antibiotics resulted in significant antagonistic interactions with inhibitory concentrations higher than when caffeine was combined with the antibiotics at the ½ MICs and MICs as shown in Table 7.

DISCUSSION
It is a common practice to wash down drugs and orally administer medications with caffeine containing soft drinks such as cola and energy drinks.Occasionally, caffeine containing food such as cocoa, coffee and tea is used in beverages to reduce the after taste some bitter drugs would leave after an oral administration.Some of these caffeine containing foods are used to persuade little children to take bitter drugs.In pharmaceutical drugs such as ergotamine, caffeine is included for the treatment of migraine headaches and is combined with nonsteroidal anti-inflammatory drugs for relieving pain (Saw-  , 1995).These practices have continued for ages without exceptional consideration for possible drug-drug or drug-food interactions between the caffeine contained by the food and the different medications, especially antibiotics used in treating bacterial infections.Therefore, it becomes necessary to investigate interactions between caffeine and antibiotics (Kang et al., 2012).
Although several reports showed that caffeine possesses antibacterial properties, most of these studies focused on caffeine containing food.For instance, Ramanaviciene and coworkers (2002) reported antibacterial activity of caffeine against Ps.flourescens and E. coli.Mohammed and Al-Bayati (2009) reported antibacterial activity of caffeine from Coffee arabica (coffee beans) and Camelia sinensis (green tea leaves) against S. aureus, B. cereus, E. coli, P. mirabilis and K. pneumoniae.Inhibitory effects of Coffea canephora extracts against L. pneumophila (Furuhata et al., 2002), S. mercescens and E. cloacae (Almeida et al., 2006) and S. mutans (Antonio et al., 2010;Almeida et al., 2012) were also reported while Suárez-Quiroz and coworkers (2004), Wilmot (2006) and Chen and coworkers (2013) showed that caffeine inhibits mold and fungi.However, there is the dearth of information on the antibacterial activities of pure caffeine powder.In this study, the minimum inhibitory concentration of pure caffeine powder ranged between 67.19 and 268.75 µg/ml.While most of the bacterial isolates were inhibited at 268.75 µg/ml, P. shigellosis ATCC 379003 and Ps.aeruginosa ATCC 15442 were inhibited at 134.38 µg/ml and E. cloacae ATCC 13047 was inhibited at 67.19 µg/ ml.These results are in agreement with previous reports.Mohammed and Al-Bayati (2009) reported that MICs of caffeine coffee ranged between 62.5 and 250 µg/ml and green tea caffeine ranged from 62.5 to 500 µg/ml while Pruthviraj and coworkers (2011) reported MICs of caffeine isolated from tea between 65.5 and 250 µg/ml and green tea caffeine between 65.5 and 500 µg/ml.The activity of pure caffeine, used in this study, was further corroborated by Ramanaviciene and coworkers (2003b) who indicated that pure caffeine has a direct antibacterial effect.
Furthermore, the interactions of caffeine with antibiotics were previously reported.While Charles and Rawal (1979) indicated that caffeine decreased the antibacterial activity of chloramphenicol and tetracycline hydrochloride, Hosseinzadeh and coworkers (2006) reported that caffeine works synergistically with carbenicillin, ceftizoxime and gentamicin which are effective against Ps.aeruginosa and S. aureus.Esinome and coworkers (2008) reported that caffeine combined with ampicillin resulted in indifferent interaction while its combination with benzypenicillin was antagonistic.Kang and coworkers (2012) reported that there are no synergistic interactions between caffeine and bleomycin or cisplatin but the activities of ciprofloxacin were decreased when paired with caffeine.In this study, the combination of caffeine with chloramphenicol, nalidixic acid, erythromycin, tetracycline and metronidazole resulted, mostly, in antagonistic interactions with the exception of the interaction with kanamycin that resulted mostly in either additive or indifferent effects.Kang and coworkers (2012), however, indicated that kanamycin activity is increased upon the addition of 0.25 mg/ml (0.025% w/v) of caffeine in a previous study.
Caffeine increases the susceptibility of bacteria and higher cells to different antibiotics and DNA-damaging agents when used in pre-treatment (Grigg et al., 1985;Petru et al. 1990;Selby and Sancar, 1990) by intercalating into the DNA (Tornaletti et al., 1989).Although caffeine interacts with bacterial nucleic acids (Sacks & Thompson, 1977;Kawamukai et al., 1986), inhibits DNA synthesis (Sandlie et al., 1983;Osman & McCready, 1998) and cause frameshift mutations (Pons & Muller, 1990), it can also interact with the enzymes responsible for repair of bacterial DNA damage by inhibiting ATP-dependent enzymes (Selby & Sancar, 1990) and inactivate ataxia-telangiectasia-mutated (ATM) and ATM-and-Rad3-related proteins responsible for the genome stability (Cortez, 2003).While the variety of the effects of combining caffeine and these antibiotics may resulted from the influence of the physicochemical interactions between caffeine and the drugs in vitro, the interactions could have resulted from the formation of complexes between the caffeine and the antibiotics (Veselkov et al., 2002).The complex formation with caffeine through hydrophobic van der Waals interactions (Lachman & Ravin, 1959;Matha & adbels, 1982;Kapuscinski & Kimmel 1993;Larsen et al., 1996) may reduce the in vitro antibacterial activity of the less polar antibiotics.The antagonistic interaction or reduction in the antibacterial activities of most of these antibiotics could also be a result of caffeine and antibiotic molecules competing for the same binding sites on DNA or "intercepting" or "protecting" properties of caffeine (Traganos et al., 1991;Davies et al., 2001) or its ability to stimulate hydrophobic molecules into their dimeric structures (Banerjee et al., 2012).Aminoglycoside antibiotics, such as kanamycin, target the ribosomes and interfere with the fidelity of protein synthesis (Poehlsgaard & Douthwaite, 2005).The interaction between caffeine and kanamycin could result from the complementary activity of both agents.Caffeine could increase susceptibility to kanamycin and intercalate into DNA (Tornaletti et al., 1989) while kanamycin induces mistranslation of mRNA to protein (Misumi & Tanaka, 1980) and damages DNA base pairs (Kang et al., 2012).On the other hand, caffeine could increase susceptibility to erythromycin but the putative complexes formed between caffeine and erythromycin possibly prevented both antibacterial agents from reaching their target sites of action.

CONCLUSION
In conclusion, caffeine demonstrated significant antibacterial activity against the selected bacterial isolates.However, its combination with the selected antibiotics resulted in significant antagonistic interactions against the bacterial isolates indicating that, against usual practice of using caffeine containing food or drinks in oral drug administration, caffeine should not be combined with antibiotics as this could result in serious therapeutic failure and, possibly, drug toxicity in vivo.