One hundred P. acnes isolates recovered from ocular clinical specimens (conjunctival swab - 65, conjunctival scraping – 1, corneal scraping – 13, lid abscess – 3, iris tissue - 2, corneal button-5, vitreous chamber tap – 1, anterior chamber tap - 2, eviscerated material – 6, orbital implant – 1 and sclera nodule - 1) received at L & T Microbiology Research centre, Sankara Nethralaya, a tertiary eye care centre at Chennai, India along with 3 American Type Culture Collection (ATCC; Manassas, Va.) (Propionibacterium acnes ATCC 11828, Clostridium sporogenes ATCC 11437 and Bacteroides fragilis ATCC 23745) were included for this study. The institutional research sub-committee approved this study protocol. Clinical specimens were processed as described elsewhere.¹²

Ocular anaerobic culture isolates received from individual patients proven to be P. acnes by conventional methods including Gram staining, catalase, indole positivity, growth characteristics in Brucella blood agar enriched with 5% sheep blood, 5 mg/L haemin and 1 mg/L vitamin K1, (Cat No. M 1039, Hi-Media Laboratories Private Limited, Mumbai) were studied. The isolates were maintained at anaerobic work station (Don Whitley Scientific Limited, Shipley, West Yorkshire, UK) (85% N₂, 10% H₂, 5% CO₂) through out the study period. The growth on Brucella blood agar was observed to be small, round, yellow to orange colored shiny and convex spherical beta haemolytic colonies.

To optimize the RAPD fingerprinting technique, the optimal concentrations of primers, DNA template, and MgCl₂ used in PCR were first determined. Trying different concentrations ranging from 15, 30, 60 and 100ng/l the primer concentration was subjected for optimization. Primer dimer formation was seen with the concentrations of 60 and 100ng/l. With 15 ng/l primer concentration the bands were very faint. Primer concentration of 30ng/l yielded clear banding patterns. Concentration of Mgcl₂ ranging from 1 to 4mM MgCl₂ produced identical banding pattern, higher concentrations of MgCl₂ yielded some artificial background, and lower concentrations of MgCl₂ resulted in poor amplification. A concentration of 2mM MgCl₂ was chosen for the RAPD reaction. Likewise, low levels of DNA template (5, 10 ng) were found to result in relatively poor amplification while higher level yielded smearing patterns. RAPD patterns, however, standardized with 20ng/ul of total DNA were used. Consequently, a slight difference in DNA concentration obtained from preparation to preparation did not affect RAPD patterns. ^{13, 14}

To select suitable RAPD primers for subtyping of 100 P. acnes isolates, 10 arbitrary primers were first tested with 10 isolates of P. acnes and one primer , 7 ^th primer (Accession no for 7 ^th primer is AM773318, Merck, Darmstadt, Germany) was found to produce three or more polymorphic patterns when tested, was taken for the RAPD determination. ^{15, 16}

DNA was extracted from a single colony isolated from Brucella blood agar using the Qiagen DNA Mini kit (Qiagen, Hilden, Germany) as per the instructions provided in the manual. The resulting genomic DNA was stored at -20°C until further subjected for PCR-RAPD analysis. PCR-RAPD was carried out for all P. acnes isolates by using 20 ng/l of genomic DNA added to 20l of PCR mixture consisting of 5μl 10X PCR buffer containing 15mM MgCl_2, 200μM dNTPs, 100ng / l of primer, 10μl deionized water of and 1.0 U of Taq polymerase (Merck, Darmstadt, Germany). PCR amplifications were carried out using thermal cycler Perkin Elmer Model 2700 (Applied Biosystems, Massachusetts, USA). PCR thermal profile consisted of initial denaturation at 94°C for 5 min, followed by 8 cycles consisting of denaturation at 94°C for 45sec, annealing at 35°C for 1 min, and extension at 72°C for 1.5 min, followed by 35 cycles consisting of denaturation at 94°C for 45sec, annealing at 38°C for 1 min, and extension at 72°C for 1 min and a final extension at 72°C for 10 min.

Entire RAPD products were loaded into 1% agarose gel (SRL, India) containing 1x Tris-acetate-EDTA buffer. Molecular size markers (100bp and 500 bp DNA ladder; (Merck, Darmstadt, Germany)) were run in parallel on all gels. Resolved DNA products were stained with ethidium bromide, 50ng/ml (Hi-Media Laboratories Private Limited, Mumbai.) and viewed under UV light using gel documentation system (VILBER LOURMAT – FRANCE). ^{15, 16}

The following antibiotics powders were used and obtained from the sources indicated: ciprofloxacin, norfloxacin, nalidixic acid, clindamycin, penicillin G, vancomycin and metronidazole (Sigma, USA), cefotaxime and imepenum (Ranbaxy, India) were used to make stock solutions that were filter sterilized and then stored at 4ºC. Final working solutions tested for ciprofloxacin, norfloxacin, nalidixic acid and vancomycin were of 0.125, 0.25, 0.5, 1, 2 and 4 mg/L respectively. For clindamycin, penicillin G, cefotaxime and imipenum, the final concentrations tested were 0.06, 0.125, 0.25, 0.5, 1 and 2 mg/L respectively. Metronidazole final concentrations were 8, 16, 32, 64, 128 and 256 mg/L respectively. ^{17 - 21}

MICs of nine antimicrobial agents were determined by agar dilution technique as described by the Clinical Laboratory Standard Institute (CLSI) standard with 10⁵ CFU⁄ spot and Brucella base sheep blood agar.^{20, 21} The plates were incubated in Don Whitley anaerobic work station (Shipley, West Yorkshire, UK) for 48 h at 37^ºC. The MIC was defined as the lowest concentration of antimicrobial agent that resulted in a marked change in the appearance of growth in comparison with the control plate, as described in the NCCLS protocol. P. acnes ATCC 11828 was used as control strain.

PCR-RAPD analysis performed with 100 P. acnes isolates with the 7 ^th set of RAPD primers showed 6 different RAPD patterns distributed among P. acnes positive ocular isolates is shown in and figure 1.

Pattern 1 was exhibited by 3 (3%) isolates of P. acnes [conjunctival swabs - 2 and eviscerated material – 1]. The products had 6 bands measuring 2500, 1500, 900, 800, 400, 100bp in size. Pattern 2 was exhibited by 18 (18%) isolates [anterior chamber tap – 1, conjunctival scraping – 1, conjunctival swab – 14, iris tissue – 1 and corneal scraping – 1]. There were totally 7 Bands with size measuring 2000, 1200, 900, 500, 400, 300, 100 base pairs. Pattern 3 was exhibited by 16 (16%) isolates [conjunctival swab - 12, corneal button – 1, corneal scraping – 2 and eviscerated material - 1]. The products produced 7 bands measuring 2000, 1200, 900, 600, 500, 400, 100bp in size. Pattern 4 was shown by 41 (41%) isolates [anterior chamber tap - 1, conjunctival swab - 22, corneal button - 4, corneal scraping – 7, eviscerated material – 2, iris tissue – 1, lid abscess – 3, sclera nodule - 1], with 5 products of size 2000, 1200, 900, 300, 100bp respectively. Pattern 5 was exhibited by 14 (14%) isolates. [Conjunctival swab - 11, corneal scraping – 2, lid abscess – 3, vitreous chamber tap – 1] There were totally 7 Bands measuring 1200, 1000, 800, 600, 500, 400, 100bp in size. Pattern 6 was shown by 8 (8%) isolates [conjunctival swab - 4, corneal scraping – 1, eviscerated material – 2 and orbital implant – 1] with 7 bands corresponding to 3000, 1500, 800, 600, 500, 300, 100bp in size respectively.

The 4 ^th pattern was found to be the most prevalent pattern among P.acnes. These different patterns from different specimens indicate that all are not of common origin or hospital acquired infection.

The reproducibility of the RAPD fingerprinting technique was confirmed by comparing the reproducibility of the fingerprint patterns obtained from duplicate runs of RAPD analysis of several different P. acnes isolates. A single primer (7 ^th primer) was used to discriminate P. acnes species based on their sequence variation. The experiments were carried out twice using two different brands of thermal cyclers, and the results were resolved on the same 1% agarose gel to evaluate the reproducibility.

Antibiotic susceptibility pattern and MIC values of 100 P. acnes isolates for the 9 antibiotics tested are shown in Table 1. The MICs for control strains were always within recommended limits. Figure 2 represents the graph showing the results of antibiotic resistance pattern prevalent among ocular P. acnes.

Of the clinical isolates, 41 isolates (41.0%) were resistant to Norfloxacin. The most resistant isolates were associated with conjunctival swabs (80%) followed by eviscerated material (80%) and corneal scraping (80%). The MICs for isolates ranged between <0.1 to 4mg/L.

Of the clinical isolates, 64 isolates (64.0%) were resistant to penicillin G. The most resistant isolates were associated with conjunctival swabs. The degrees of resistant isolates were high with conjunctival swabs (40%) followed by corneal scrapping (33%) and eviscerated material (75%). The MICs for isolates ranged between <0.1 to 2mg/L.

A total of 61 (61.0%) clinical isolates were resistant to imipenum. The most resistant isolates were associated with conjunctival swabs (42%) with the MICs ranging between <0.1 to 2mg/L.

Of the clinical isolates, 50 isolates (50.0%) were resistant to vancomycin. The most resistant isolates were associated with conjunctival swabs followed by eviscerated material and corneal scrapping with MICs ranging from between <0.1 to 4mg/L.

Of the clinical isolates, 42 isolates (42.0%) were resistant to nalidixic acid. The most resistant isolates were associated with conjunctival swabs. The MICs were ranging between <0.1 to 4mg/L.

A total of 79 (79.0%) clinical isolates were resistant to clindamycin. Resistance to clindamycin was found to be predominant among conjunctival swabs isolates with MICs ranging from between <0.1 to 2mg/L.

All the isolates were susceptible to ciprofloxacin.

All the isolates were susceptible to cephotaxime.

All P. acnes strains (100%) were resistant to metronidazole with an MIC >256 mg/L.

There existed no correlation between specific banding patterns of RAPD with that of the MIC results. Higher resistance was seen among the extra ocular isolates especially the conjunctival swabs and however predominantly many conjunctival swabs exhibited pattern 4.

Figure 1

Figure 1: Agarose Gel electrophoretogram of six different RAPD finger printing profiles of the using 7 primer set.

Pattern 1 - 6 bands measuring 2500, 1500, 900, 800, 400, 100bp in size.

Pattern 2 - 7 Bands measuring 2000, 1200, 900, 500, 400, 300, 100bp in size.

Pattern 3 –7 Bands measuring 2000, 1200, 900, 600, 500, 400, 100bp in size

Pattern 4 – 5 Bands measuring 2000, 1200, 900, 300, 100bp in size

Pattern 5 – 7 Bands measuring 1200, 1000, 800, 600, 500, 400, 100bp in size

Pattern 6 - 7 Bands measuring 3000, 1500, 800, 600, 500, 300, 100bp in size

Figure 2

Figure 2: Graph showing the results of antibiotic resistance pattern prevalent among ocular

Figure 3

Table 1: Results of In-vitro susceptibility of 100 isolates

Performed against nine antimicrobial agents