Biofilm production and antibiotic susceptibility profiles of Staphylococcus aureus isolated from HIV and AIDS patients in the Limpopo Province , South Africa

Staphylococcus aureus is a common pathogen associated with nosocomial as well as community acquired infections. Despite multiple reports on the severity and recurrent nature of S. aureus infection, the pathogenesis as well as antibiotic susceptibility profiles of S. aureus infecting HIV and AIDS patients has not been well studied in Limpopo Province, South Africa. Hence, the study was aimed at determining the biofilm producing capability and antibiotic resistance profiles of the S. aureus isolated from drinking water and clinical samples from HIV patients in the Limpopo Province. S. aureus strains isolated from clinical samples including urine, sputum, and stools obtained from HIV and AIDS patients as well as their drinking water samples were analysed for biofilm production using the microtitration plate method, haemolytic activities, antibiotic susceptibility, methicillin resistance and β-lactamase production. Methicillin resistance was tested from all the S. aureus isolates by the oxacillin agar plate. Overall, 140 S. aureus were isolated. Sixty (60) were from stools, 48 from urine, 15 from sputum and 17 from water. Of all the isolates, 67 (48%) produced biofilm of which 14 (10%) were strong biofilm producers and 53 (38%) moderate biofilm producers. Biofilm production was the highest among the urine isolates (62.5%) (χ 2 = 6.276; p = 0.01) and the water isolates (70.6%) (χ 2 = 4.006; p = 0.040). However, biofilm was not associated with sputum samples (p = 0.571) nor stool samples (p = 0.763). Isolates were highly resistant to most antibiotics tested and 125 (90%) isolates were resistant to more than 3 antibiotics. Higher resistance was observed against ampicillin (92%) while the most active antibiotic was ciprofloxacin with 89% susceptibility. Eighteen (14%) of the isolates were identified as MRSA and showed high resistance against vancomycin (28%) compared to MSSA among which vancomycin resistance was 14%. This is the first study on staphylococcal isolates from HIV patients in the Limpopo Province. The level of resistance to vancomycin was high and further attention is needed from the health system for more stringent measures of infection control. Biofilm production appeared to be a factor fuelling the increased in antibiotic resistance as well as pathogenicity in these strains. Furthermore, water could be a transmission vector of staphylococcal UTIs among HIV and AIDS patients in this region. However, further studies are needed to confirm these hypotheses.


INTRODUCTION
Human immunodeficiency virus (HIV) and acquired *Corresponding author.E-mail: samieamidou@yahoo.com.immunodeficiency syndrome (AIDS) is a major health problem in South Africa and HIV and AIDS patients are vulnerable to several infections termed opportunistic infections because of their weakened immune system.Over the past two decades, Staphylococcus aureus has emerged as a significant opportunistic pathogen among HIV and AIDS patients in both nosocomial and community settings, and recent studies have shown greater frequency and morbidity of this organism among HIV positive individuals (Chacko et al., 2009;Hidron et al., 2010).However, there is little data on the occurrence, pathogenicity and antibiotic resistance of this organism among HIV positive patients in the Limpopo Province, South Africa.

Abbreviations
S. aureus normally localize in the skin and mucous membranes in the nose of healthy humans and about 30% of the normal healthy population are transiently colonised by the organism (Liu, 2009) which has been associated with several syndromes such as skin infections, osteomyelitis, bacteraemia, septicaemia, diarrhea, pneumonia and urinary tract infections (Tumbarello et al., 1996;Franzetti et al., 2006;Flemming and Ackermann, 2007;Muder et al., 2006;Baba-Moussa et al., 2007;Shigemura et al., 2005;Huggan et al., 2008).These syndromes have been mostly described among HIV and AIDS patients and are responsible for high rates of morbidity and mortality.In South Africa, few literatures exist on the different types of S. aureus infections, particularly among HIV positive patients.Developing countries are at risk of getting waterborne disease as a result of consuming drinking water which is contaminated with pathogenic microbes of faecal origin.Death resulting from water-borne diarrhoeal disease is estimated at over 3 million annually, especially among infants and young children in poor communities in Africa, Asia and South America (Battu and Reddy, 2009).Normally, S. aureus should not be present in drinking water, but many studies have isolated high rates of S. aureus in water (Mihdhair, 2009).Residential care centres, communities and hospitals are often supplied with contaminated water and results in the distribution of infection among patients, particularly those with weaken immune system (Johnson et al., 2009).
Biofilm formation has been described as a possible attribute to the resistance and major contributor to nosocomial infections.Biofilm impairs the action of both host immune system and antimicrobial activities.Biofilm production is an important virulence factor of S. aureus (Dhanawade et al., 2009).The formation of biofilms is an example of a phenotypic change in S. aureus to adapt to its surroundings in the presence of environmental challenges (Jain and Agarwal, 2009), and is a recognized method of some organisms' ability to establish and maintain certain infections, and a way which increases its persistence and boosts its levels of antimicrobial resistance (Monre, 2007).Diseases such as endocarditis, osteomyelitis and medical-device related infections are caused by S. aureus biofilms and are not readily treatable with antibiotics.
Although, there is limited data on the prevalence of staphylococcal infection in Africa, one of the earliest case of MRSA was reported in South Africa (Shittu and Lin, 2006).Baseline data from Pan-European antimicrobial resistance using local surveillance (PEARLS) study of 2001 to 2002 showed that South Africa had MRSA prevalence of 33.3% with 23% MRSA in patient around Johannesburg (Marais et al., 2009).The prevalence of MRSA was lower, about 10% in Tunisia, Malta, and Algeria and around 15% in Kenya, as compared to the high prevalence of 21 to 30% in Cameroon and Nigeria when many African hospitals were studied (Kesah et al., 2003).In Uganda, about 10% of the surgical procedures become septic which account for an increasing morbidity and mortality with the commonest organism isolated being S. aureus (Ojulong et al., 2009).Data from the national nosocomial infection surveillance network (USA) have shown that MRSA present more than 50% of S. aureus strains causing infection in patients in intensive care units (Hidron et al., 2005).In Europe, MRSA prevalence ranges from over 50% in Portugal and Italy to below 2% in Switzerland and the Netherlands, where infection control measures have been shown to work (Ojulong et al., 2009).In Asia, the prevalence lies around 50%, with extremely high rates in Hong Kong (75%) and Japan (70%) (Saikia et al., 2009).The prevalence of MRSA in the Northern part of South Africa however, is not known.
Antimicrobial resistance is an increasing problem and a challenge worldwide (Uneke et al., 2010).The prevalence of resistance in S. aureus also is increasing globally.Staphylococcal resistance to penicillin is mediated by penicillinase production: an enzyme which breaks down the β-lactam ring of the penicillin molecule (Bassetti et al., 2009).The identification of possible associations between biofilm production and pathogenesis as well as antibiotic susceptibility profiles of infecting S. aureus could provide better control measures particularly among immunocompromised individuals.This study determined the prevalence of biofilm formation among S. aureus isolates from clinical and water samples in relation to antibiotic susceptibility and haemolytic activity.

Ethical considerations
Ethical clearance of the study was obtained from the University of Venda Health Safety and Ethics Committee.Authorization to conduct the study was obtained from the Department of Health, Limpopo in Polokwane.Ethical clearance and authorization was also obtained from the ethical committees of the Donald Fraser and Tshilidzini Hospitals.The objectives of the study were explained to the patients in their mother tongue (Tshi-Venda, Tsi-Pedi or Tsi-Tsonga) and their right to say no to participate in this study was explained to them.Once the patients had agreed to participate in the study they were requested to sign a consent form.To preserve their privacy, the patients were given a code and were referred to by that code.As such, different samples including sputum, urine, mouth wash and stools were collected whenever it was possible to do so.Patients with HIV positive, visiting major hospitals in the Vhembe district, situated in the Northern part of South Africa.

Bacterial isolation and identification
The different samples were transported to the laboratory within 4 h of collection and were inoculated onto freshly prepared mannitol salt agar and incubated at 37°C for 24 h.Golden yellow colonies were presumptively identified as S. aureus.Then a single golden yellow colony from the plate was subcultured on freshly prepared nutrient agar and incubated at 37°C for 24 h.The nutrient agar plates were then stored in the fridge for further analysis.As a result, a total of 140 isolates were obtained and used in this study.

Confirmation of the isolates by polymerase chain reaction (PCR)
Genomic DNA was extracted from the isolates using the protocol described by Kumar et al. (2008) and the quality of the DNA was checked by agarose gel electrophoresis.The DNA was stored in the freezer at -20°C until further analysis.
The identification of S. aureus by polymerase chain reaction was done through the detection of tuf-gene from the genomic DNA extracted from the isolates according to Jimenez et al. (2008) with some slight modifications.The primer set used included tuf-g (5´-GGTGTACCAGCATTAGT-3´), tuf-e (5´-TTCGTGCATACCGATGA) and tuf-a (5´TTCAGTATGTGGTGTAA-3´).The reaction was done in a multiplex PCR format to allow the identification of Staphylococcus epidermidis strain from isolates if present.The specificity of the assay was confirmed using a negative control of 5 µl of nuclease free water.The PCR amplicons were detected using 1% agarose gel electrophoresis stained with ethydium bromide.

Haemolytic activity
The haemolytic activity testing of the S. aureus isolates was performed according to the method previously described by Jimenez et al. (2008).Briefly, Colombia agar was prepared and supplemented with 5% horse blood.The microorganisms were inoculated on Columbia blood agar and incubated at 37°C.The plates were analyzed after 72 h and isolates were classified as nonhaemolytic when no lysis was observed, beta (β) hemolytic when moderate lysis was observed and alpha (α) hemolytic when complete lysis of the blood in the media was observed.For better visibility, about 8 microorganisms were tested in a plate.

Biofilm assay
In order to test for biofilm production by the different strains of S. aureus isolated from clinical and water samples, two different methods were used namely; the microtitre plate method (Kwon et al., 2008) and the Congo red agar based method (Freeman et al., 1989).

Microtitre plate method
The assay was performed using U shaped polystyrene microtitre plates with brain heart infusion broth (BHI) supplemented with 1% sucrose (Kwon et al., 2008).Prior to inoculation on the plates the microorganisms were grown in BHI for 2 to 4 h.199 µl of the 1% sucrose BHI was added to each well of the microtitre plate and 1 µl of the microbial suspension was added.All the reactions were done in triplicate (that is, using three wells for each strain).The plate was covered and incubated at 37°C for 24 h.The next day, the plates containing the culture were emptied and washed with phosphate buffered saline (PBS) five times to remove any unfixed microbial cell and leave only those fixed in the well within a biofilm matrice.Then 175 µl of 1% crystal violet was added and incubated at room temperature for 15 min.

Samie and Shivambu 14627
The plates were further washed 5 times with PBS and dried for 30 min at room temperature.200 µl of ethanol-acetone (80 to 20%) were added to solubilise the microorganism and the crystal violet and incubated at room temp for 25 min.The optical density was recorded using an ELISA reader (Bio-Tek Instruments, INC Winooski, VT, USA) at the wavelength of 490 nm.The strains that gave an OD value less than 0.120 were recorded as non-biofilm producers, those with OD values between 0.120 and 0.240 were considered as moderate producers and those with OD values more than 0.240 were considered as strong producers.The reaction was repeated at two different occasions.

The Congo red agar based method
The agar medium used was prepared by adding 37 g of the BHI powder, 50 g of sucrose and 10 g of agar No 1 in 1 L of distilled water.The mixture was then autoclaved for 15 min at 121°C.Once the agar solution has cooled down to about 50°C, a solution of Congo red (8 g/L) was added and mixed again and then the media was poured into the Petri plates and allowed to solidify.Once the media had settled, the plates were inoculated with the microorganisms and incubated at 37°C for 24 h.The plates were read the next day and the organisms were considered positive (biofilm-producers) when they produced black colonies on the agar and negative (non-biofilm producers) when they produced pink, or red-orange colonies on the Congo red agar (Freeman et al., 1989).

Antibiotic susceptibility profiles of Staphylococcus aureus
The disc diffusion method was used in order to determine the antibiotic susceptibility profiles of the strains.The microorganisms were prepared in Mueller-Hinton broth (MHB) by directly suspending one colony in a tube of sterile MHB.Prior to testing for antibiotic susceptibility, a 0.5 McFarland standard of the organism was prepared in MHB.The test was conducted on Mueller Hinton agar plates using a swab and the 0.5 MacFarland culture.The cultures were inoculated into the plates using sterile swabs and left for about 10 min and the antibiotics discs were placed on the culture and incubated at 35 to 36.5°C for 24 h.The zone of inhibition was measured the following day and the results were interpreted according to CLSI (previously known as NCCL) standards (CLSI, 2007).The antibiotics tested included amoxycillin (AML 10), ampicillin (AMP 10), cefepime (FEP 30), ceftriaxone (CRO 30), cefoxitin (FOX 30), chloramphenicol (C 30), erythromycin (E 15), kanamycin (K 30), meropenem (MEM 10), penicillin G (P 10), polymyxin B (PB 300), streptomycin (S 10), tetracycline (30) and vancomycin (VA 30).

Detection of methicillin-resistant Staphylococcus aureus by the agar method
Mannitol salt agar was prepared and cooled to about 55°C.The media was supplemented with 2 mg/L of oxacillin.Microorganisms were inoculated into the oxacillin agar plate and incubated at 37°C for 42 h.Growth of isolates was identified as positive.Eight isolates were tested per plate (Arora et al., 2010).

Statistical analysis
All data was entered into a Microsoft excel sheet.The analysis was conducted using the statistical package for social sciences (SPSS) program, version 17.1.Chi-square test was used for comparison of the different variables and the correlation between all the tests performed.A p value of <0.05 was considered to be statistically significant.

Bacterial isolates used in this study
A total of 140 strains of S. aureus were isolated from different types of samples (from a total of about 650 samples).All the isolates were confirmed to be S. aureus by PCR by the observation of the specific bands as shown in Figure 1.Of these, 48 (34%) were from urine, 60 (43%) from stools and 15 (11%) from sputum.The remaining 17 (12%) isolates were from water samples.Of all the isolates, 79 (64%) were from females, while 44 isolates were from males.The age of the patients varied between 18 and 81 years.
There was no significant difference between sample type and haemolytic activity, (p>0.05).

Biofilm production
The microtitre plate method (MTP) was compared to the Congo red agar based method for the detection of biofilm production.It was easy to discriminate between strong biofilm producers, moderate and non producers when using the MTP (Figures 2A), while the CRA could only identify microorganism as producers and non producers (Figures 2B).There was a good correlation between the two methods, (p<0.05),but the MTP methods was the one used for further analysis, because it was found to be more sensitive and easy to read, although, both methods were easy to perform.With the polystyrene microtitre plate, 14 (10%) were strong producers, 53 (38%) were moderate biofilm producers while 73 (52%) were non biofilm producers.With the Congo red methods, 53 (37.3%) biofilm producers and 87 (62.1%) were nonbiofilm producers.Figure 3 shows a scatter plot of the OD values obtained from the biofilm producers.The microtitre plate method was more sensitive as compared to the Congo red agar method, and it was the one used for further analysis in this study.Organisms from water samples, followed by those from urine samples had the highest rate of biofilm production.However, water samples had the highest percentage of strong producers (17.6%), followed by those from stools (10%) compared to the isolates from other samples.Most organisms from urine samples showed moderate biofilm production phenotype (56.3%) (Table 2).There was a significant difference between sample origin and biofilm production (p<0.05).

Antibiotic susceptibility/resistance profiles
High resistance was observed against the penicillin group and cephalosporines, while resistance was observed against the carbapenems, fluoroquinolones and glycolpeptides groups.None of the isolates was completely susceptible to all the tested antibiotics.Resistance to ampicillin, cefepime, tetracycline and erythromycin were 92, 76, 63.5 and 52% respectively.In contrast, there was less resistance to meropenem (18.8%), chloramphenicol (14.5%), vancomycin (14.5%) and ciprofloxacin (11.4%).Table 3 summarizes the antibiotic susceptibility profiles of the isolates against 14 antimicrobial agents.

Antibiotic resistance and sample types
Isolates from all type of samples showed high resistance to the penicillin group with 100% resistance to ampicillin observed from sputum and water isolates.Within the aminoglycosides group, 32.8% of urine isolates were resistant to kanamycin as compared to only 6.7% of the sputum isolates.Twenty nine percent of urine isolates were resistant to streptomycin, while all water isolates were susceptible (0% resistance) to the same antibiotic.Furthermore, both meropenem and chloramphenicol were 100% active against the water and sputum isolates, although, 27.1 and 26.4% resistance were observed among the urine and stool isolates respectively.Vancomycin was 100% active against sputum isolates with low resistance in the urine isolates (6.3%), and high resistance 26.9% in the stool isolates (Figure 3A and B).

Multiple drug resistance among the isolates obtained from clinical and water samples from HIV positive patients in the Limpopo Province
Multiple drug resistance (MDR) was defined as resistance to 3 or more antibiotics (ATB).From 140 isolates tested against 14 antibiotics, 125 (90%) of the isolates were resistant to at least 3 or more ATB.None of the isolate was susceptible to all antibiotics.The highest rate of multiple drug resistance was observed against 7 antibiotics with 26 (18.7%) isolates, followed by isolates resistant to 6 antibiotics 24 (17.3%), while 16 (11.5%)were resistant to 8 antibiotics at a time.None of the isolates were resistant to 13 and14 antibiotics, while 1 (0.7%) was resistant to 12 antibiotics.

Multiple drug resistance and sample type
The occurrence of multiple drug resistance among the isolates was compared between sample types.From those that resisted to the 7 antibiotics, higher resistance rate was observed among the water isolates (29.4%) as compared to 13% among sputum.Four (23.5%) of the water isolates were resistant to 9 antibiotics with only 6.25% among urine isolates and none among sputum.Six (12.5%) of the urine isolates were resistant to 10 antibiotics, while none of the sputum and water isolate was resistant to 10 antibiotics at a time (Figure 4).

Antibiotic resistance and gender
The rate of penicillin resistance among female was higher compared to the resistance rate in males (86 and 81% respectively), however, the difference was not statistically significant (p>0.05).The resistance profile to most antibiotics used was generally similar among females and males, although, slight differences were observed with the aminoglycoside, 26% of the isolate from females were resistant to kanamycin compared to 30% for the isolates obtained from males, while 20% females' isolates were resistant to streptomycin compared to 12% of the males.A difference was observed with the vancomycin in which 23% isolates from female displayed resistance with only 6% among the strains isolated from male patients.

Multiple drug resistance and gender
60 (92%) of the isolates from female were multi-drug resistant as compared to 28 (84.8%) in males.Among the isolates from female, higher rate of MDR was observed against 6 ATB (23%); while only 9% was observe among the strains isolated from males against the same number of antibiotics.In some cases, there was high rate of MDR by male isolates as compared to female isolates.While the rate of MDR to 4 and 8 antibiotics was 18 and 15.1% respectively among the male isolates, the rate was only 9.2 and 9.2% for the female isolates.None of the male isolates were resistant to 11 antibiotics at a time, while 4.6% of female isolates displayed resistance to 11 ATB at a time.

Comparison of the two methods
All samples were tested for the β-lactamase activity using the two methods.With the tube method, 55 (39.6%) were β-lactamase positive, while 84 (61.4%) were β-lactamase negative.With the agar method, 53 (38%) were positive and 86 (62%) were negative.Among the 55 positive with the tube, 40 were also positive with the agar while the rest were negative.The tube method was the one selected and used for further analysis.

Association between β-lactamase and resistance to other antibiotics
β-Lactamase was associated with resistance of isolate to other antibiotics.β-Lactamase positive isolates showed 92 and 94% resistance to penicillin and ampicillin   respectively.β-Lactamase positive isolates showed least association with resistance to chloramphenicol (14.5%).Β-Lactamase was significantly associated with resistance to ciprofloxacin (χ 2 = 7.383; p = 0.008).Although, vancomycin resistance was higher among beta lactamase positive isolates, the difference was not statistically significant (p = 0.224).

β-lactamase and multiple drug resistance
β-Lactamase production of the isolates was also compared with multiple drug resistance.Highest rate of MDR was observed among the β-lactamase producing isolates (93%) compared to 88% among the β-lactamase negative isolates.High β-lactamase MDR to 7 antibiotics was observed among positive isolates (21.1%) compared to 16.7% among negative isolates.The only isolate that was resistance to 12 antibiotics was a β-lactamase positive isolates (1.8%) (Table 4).β-Lactamase production rate was higher among female (51%) compared to males (26%).

Overall antibiotic susceptibility profile of the biofilm producing and non producing isolates
In general, there was no significant difference between antibiotic resistance among biofilm producing strains and non-producing strains.However, biofilm producing strains were more resistant to meropenem with 42.8% resistance compared to only 13.7% resistance among moderate biofilm producers and 17.8% resistance among nonbiofilm producing strains (χ 2 = 6.394; p = 0.042).Also, 92.8% strong producers were resistance to ceftriaxone, while 64 and 63% of moderate and non-producers showed resistance to this antibiotic (Table 5).

DISCUSSION
Over the past decade, there has been an increase in the rate of infection and diseases caused by S. aureus particularly MRSA throughout the world (Sadaka et al., 2009).The situation is even more alarming among significant difference between distribution of MRSA and sample origin (p = 0.495) except for water.
Bacteria use biofilm mechanism as a way of causing chronic infection to human (Leid et al., 2002).Biofilm are also well suited for resistance to antibiotics and evasion of immune system's defences.The development of biofilm by S. aureus might be a mode of adaptation of these organisms to live in different environments.Furthermore, biofilm-mediated infections in the hospital environment have a significant negative impact on patient's health and place an enormous burden on the resources of the health services (Smith et al., 2008).The study characterized most strong biofilm producers from water with 17.6% as compared to clinical isolates.Householders should be advised on their water storage and a call for best water treatment methods.From the clinical isolate those from stools were also strong biofilm producers (10%) as compared to sputum and urine, thus, it should be important to look at the association between drinking water, biofilm and diarrhoea which this study could not clarify.There was a significant difference between biofilm production by the clinical isolates (p<0.05) and is in contrast to previous studies (Critchley et al., 2002).Although, biofilm formation in water have been identified as a major problem in drinking water and can results in food intoxication, biofilm production from urine pose a major public health burden for people requiring indwelling medical devices.The ability of nosocomial pathogens including S. aureus, to form biofilms is of significant clinical interest, since biofilm formation influences the efficacy of antimicrobial therapy, the subsequent outcome of an infection, increased prevalence of antibiotic resistance and induce resistance also to vancomycin (Götz, 2002).However, in our study, there was no clear relationship between biofilm production and vancomycin although strong producers were significantly resistant to meropenem.

Conclusions
This study has shown that strong biofilm production by S. aureus strains isolated from HIV and AIDS patients was predictive of meropenem resistance and β-lactamase production.Although, we detected low MRSA compared to other province, the high antibiotic resistance and multidrug resistance observed among these isolates is of serious concern for the treatment of these infections in HIV and AIDS patients.S. aureus and particularly MRSA appear to be responsible for urinary tract infections as well as respiratory infections among HIV and AIDS patients in the Limpopo Province.Drinking water seems to be a source of transmission of pathogenic strains of S. aureus.However, the pathogenicity of these strains needs to be confirmed by using larger number of patients in comparison with HIV negative patients.Further studies are needed in order to identify the genetic features contributing to biofilm formation among these isolates for Samie and Shivambu 14635 better management of S. aureus infections among the HIV and AIDS patients and to reduce the level of morbidity and mortality among patients suffering from S. aureus infections.There is also a need to maintain surveillance control of MRSA infection in South Africa because the majority of MRSA are multidrug resistant as identified in our study and studies outside of South Africa.

Figure 1 .
Figure 1.Agarose gel electrophoresis pattern showing the 530 bp of the amplification product for S. aureus tuf-gene.Lanes: M DNA molecular Marker (1000 bp ladder: Marker X, from Invitrogen, CA, USA); lanes 1 to 4 tuf-gene positive amplicon products, NC is negative control (E. coli DNA amplification).

Figure 3 .
Figure 3. Antibiotic resistance profile of the isolates by sample type.

Figure 4 .
Figure 4. Pattern of multiple drug resistance of S. aureus by sample type.

Table 1 .
The overall distribution of haemolytic activity of the isolates by sample type.

Table 2 .
Biofilm production by organisms from different types of sample.

Table 3 .
Antibiotic resistance of the S. aureus isolates from HIV patients in the Limpopo province.

Table 4 .
Characteristics of MRSA positive samples from HIV/AIDS patients in the Limpopo Province.