Free
Immunology and Microbiology  |   August 2012
Comparative Ocular Microbial Communities in Humans with and without Blepharitis
Author Affiliations & Notes
  • Se Hee Lee
    From the School of Biological Sciences, Research Center for Biomolecules and Biosystems, Chung-Ang University, Seoul, Republic of Korea; and the
  • Doo Hwan Oh
    Department of Ophthalmology, College of Medicine, Chung-Ang University, Seoul, Republic of Korea.
  • Ji Young Jung
    From the School of Biological Sciences, Research Center for Biomolecules and Biosystems, Chung-Ang University, Seoul, Republic of Korea; and the
  • Jae Chan Kim
    Department of Ophthalmology, College of Medicine, Chung-Ang University, Seoul, Republic of Korea.
  • Che Ok Jeon
    From the School of Biological Sciences, Research Center for Biomolecules and Biosystems, Chung-Ang University, Seoul, Republic of Korea; and the
  • Corresponding author: Che Ok Jeon, School of Biological Sciences, Chung-Ang University, 84, HeukSeok-Ro, Dongjak-Gu, Seoul, 156-756, Republic of Korea; cojeon@cau.ac.kr
Investigative Ophthalmology & Visual Science August 2012, Vol.53, 5585-5593. doi:10.1167/iovs.12-9922
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Se Hee Lee, Doo Hwan Oh, Ji Young Jung, Jae Chan Kim, Che Ok Jeon; Comparative Ocular Microbial Communities in Humans with and without Blepharitis. Invest. Ophthalmol. Vis. Sci. 2012;53(9):5585-5593. doi: 10.1167/iovs.12-9922.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: The aims of our study were to compare the ocular microbial communities of humans with and without blepharitis in an attempt to elucidate which microorganisms may cause blepharitis.

Methods.: Bacterial 16S rRNA genes of eyelash and tear samples from seven blepharitis patients and four healthy controls were sequenced using a pyrosequencing method, and their bacterial community structures were compared bioinformatically.

Results.: Phylotypic analysis demonstrated that eyelash and tear samples had highly diverse bacterial communities with many previously undescribed bacteria. Bacterial communities in eyelash samples from subjects with blepharitis were less diverse than those from healthy controls, while the bacterial communities of tear subjects with blepharitis were more diverse than those of healthy subjects. Statistical analyses using UniFrac and a principle coordinate analysis showed that the bacterial communities of tear samples from subjects with blepharitis were well clustered, regardless of individual, while the bacterial communities of all eyelash samples and healthy tear samples were not well clustered due to high interpersonal variability. Bioinformatic analysis revealed that Propionibacterium, Staphylococcus, Streptophyta, Corynebacterium, and Enhydrobacter were the common ocular bacteria. An increase of Staphylococcus, Streptophyta, Corynebacterium, and Enhydrobacter, and a decrease of Propionibacterium were observed from blepharitis subjects, in terms of the relative abundances.

Conclusions.: Higher abundances of Streptophyta, Corynebacterium, and Enhydrobacter in blepharitis subjects suggested that human blepharitis might be induced by the infestations of pollens, dusts, and soil particles. These results will provide valuable information for the prevention and treatment of human blepharitis based on ocular microbial flora.

Introduction
Blepharitis is the general term used to refer to inflammation involving the eyelids and it induces various discomforts, including burning, itching, irritation, and photophobia. 1 It is one of the most commonly encountered conditions in the practice of ophthalmology. However, most of the common causes of blepharitis remain ill-defined, defying detailed pathophysiology with unknown mechanisms. Many researchers have reported that Demodex may be an important etiologic factor in chronic blepharitis, conjunctival inflammation, ocular rosacea, and meibomian gland dysfunction, and its occurrence has been shown to be related significantly to age, ocular discomfort, tear film instability, and poor ocular hygiene. 27 Until now, very little has been known of the fundamental factors of Demodex mites that cause blepharitis symptoms. There have been reports that Demodex-related Bacillus, rather than Demodex itself, might stimulate inflammatory diseases, like rosacea, on the forehead, cheek, chin, or nose due to serum immunoreactivity. 811 For example, Lacey et al. 8 and Li et al. 9 reported that 83-kDa and 62-kDa antigen proteins produced by a Demodex folliculorum-related bacterium, Bacillus oleronius , elicited host immune responses and exacerbated cutaneous inflammation by serum immunoreactivity in blepharitis patients. However, despite the high incidence of Demodex in blepharitis patients, it still is controversial as a cause, as blepharitis symptoms are not always present in patients with Demodex and often are found in individuals without Demodex. 1214  
On the other hand, many researchers have posited that very diverse microorganisms inhabit the ocular environments, and some of these microorganisms may trigger blepharitis symptoms. Therefore, many studies have been performed using culture-dependent approaches to investigate ocular microbial communities in an attempt to elucidate which microorganisms may cause blepharitis. 1519 For example, Dougherty and McCulley, 15 and Kulaçoğlu et al. 18 reported that Staphylococcus and Propionibacterium strains were identified from blepharitis patients as major isolates, and that their elevated levels might contribute to the occurrence of blepharitis. However, culture-based approaches have many limitations, including that many microorganisms are difficult to culture. 20,21 Therefore, culture-independent analysis based on 16S rRNA gene sequences has been developed and applied to investigate human body microbial communities. 22,23 Culture-independent approaches, including PCR, denatured gradient gel electrophoresis (DGGE), and pyrosequencing, have been applied to the study of ocular microbial communities and have revealed that sets of previously cultured isolates from eye environments did not reflect the true microbial compositions of ocular microbial flora. 2426 In the current study, we applied a massively parallel pyrosequencing strategy to compare more specifically the ocular microbial communities of blepharitis patients and healthy controls. These results will expand our information on the ocular microbial communities, and contribute to the prevention and treatment of blepharitis. 
Methods
Sample Collection
Seven patients with blepharitis who visited a clinic (Chung-Ang University Hospital) for ophthalmic examinations between September 1 and November 30, 2010, and four controls without blepharitis were included in the study. Informed consent was obtained from all participants. This study was approved by the Chung-Ang University Hospital Institutional Review Board, and all methods adhered to the principles of the Declaration of Helsinki. The information from all participants is summarized in Table 1. All participants underwent a complete ophthalmic examination under a slit-lamp biomicroscope. Blepharitis was diagnosed based on clinical evidence of lid margin or tarsal conjunctival erythema, bulbar conjunctival hyperemia, telangiectasia, thickening, or irregularity of the eyelid margins, or meibomian gland orifice inclusions. Eyelash samples were obtained by epilation of four eyelashes (two eyelashes from each lower and upper lids), and the number of Demodex mites was counted with an optical microscope. We attempted to epilate as deeply as possible lashes with cylindrical dandruff around the root of the lash. For tear sampling, a small amount of 0.9% (wt/vol) saline solution was dropped on the bulbar conjunctiva of both eyes, and then the participants blinked several times to spread the saline solution to the corner of the bulbar conjunctiva. Tear samples were collected from the corner of the bulbar conjunctiva using hematocrit-capillary tubes (Haematokrit-kapillaren, Eberstadt, Germany), and the presence of Demodex was determined using an optical microscope. The obtained eyelash and tear samples were transferred to 0.6 mL tubes and stored in a −80°C freezer until DNA extraction. 
Table 1. 
 
Demographics of Blepharitis Patients and Healthy Controls for Eyelash and Tear Sampling
Table 1. 
 
Demographics of Blepharitis Patients and Healthy Controls for Eyelash and Tear Sampling
Group Subject No. Age (y) Sex Demodex* Allergy†
Left Right
Blepharitis patients (B) 1 63 Female +
2 66 Male + +
3 65 Male + + +
4 66 Male + + +
5 69 Male + +
6 64 Male + +
7 76 Female
Healthy controls (H) 1 54 Male
2 76 Female
3 25 Male
4 27 Male +
DNA Extraction and PCR Amplification for Barcoded Pyrosequencing
To extract total genomic DNA from eyelash and tear samples, 0.3 g of 0.1 mm zirconia/silica beads (Biospec, Bartlesville, OK) and 50 μL of 5% (wt/vol) Chelex-100 (BioRad, Hercules, CA) were added to the 0.6 mL tubes containing eyelash or tear samples, and the tubes then were vortexed vigorously for 2 minutes. After boiling for 10 minutes, the tubes were vortexed vigorously again for 2 minutes and centrifuged for 2 minutes at a maximum speed. The supernatants of the samples were used as templates for PCR amplification of bacterial 16S rRNA genes. For barcoded pyrosequencing, bacterial 16S rRNA genes containing hypervariable regions (V1–V3) were amplified using primer sets, Bac9F (5′-adaptor B-AC-GAG TTT GAT CMT GGC TCA G-3′)/Bac541R (5′-adaptor A-X-AC-WTT ACC GCG GCT GCT GG-3′), 27,28 where X denotes unique 7–11 barcode sequences inserted between the 454 Life Sciences adaptor A sequence and the common linker, AC (see Supplemental Table S1). All PCR amplifications were done in a 50 μL C1000 thermal cycler (BioRad) containing 5 μL of template genomic DNA, 20 pmol of each primer, and a Taq polymerase mixture (Solgent, Daejeon, Korea), using a cycling regimen of 94°C for 5 minutes (1 cycle), 94°C for 45 seconds, 56°C for 45 seconds, 72°C for 1 minute (30 cycles), and 72°C for 10 minutes (1 cycle). 
Pyrosequencing and Data Analysis
The PCR products were purified using a PCR purification kit (Solgent), and their concentrations were assessed carefully using an ELISA reader equipped with a Take3 multivolume plate (SynergyMx; BioTek, Winooski, VT). A composite DNA sample was prepared by pooling equal amounts of PCR products from each sample. Pyrosequencing of the composite DNA sample was performed on 1/8 plate two times by Macrogen (Seoul, Korea) using a 454 GS-FLX Titanium system (Roche, Branford, CT). Pyrosequencing data were processed and analyzed using the RDP pyrosequencing pipeline (available in the public domain at http://pyro.cme.msu.edu). 29 The sequencing reads were assigned to specific samples based on their unique barcodes sequences, and then the barcodes were removed. The resulting sequencing reads were trimmed by removing beginning and ending bases with a quality score <20 (error rate 0.01), and only sequences >300 base pairs (bp) in length were chosen for further analyses using the Pipeline initial process. Unexpected or nonbacterial reads were removed manually using the RDP classifier. 30 Taxonomic assignments of the processed bacterial reads were performed using the RDP naive Bayesian rRNA Classifier at an 80% confidence threshold. Operational taxonomic units (OTUs) and rarefaction curves were generated using the RDP pyrosequencing pipeline at a 3% dissimilarity level. The Shannon-Weaver 31 and Chao1 biodiversity indices, 32 and evenness were calculated by the RDP pyrosequencing pipeline. The bacterial community structures of eyelash and tear samples were compared using a UniFrac analysis 33 based on the phylogenetic relationships of representative sequences derived from all reads of the individual samples. Briefly, the processed read sequences were clustered into OTUs using CD-HIT 34 with an identity cutoff of 97%. The representative sequences from CD-HIT were aligned using NAST 35 based on the greengenes database, 36 with a minimum alignment length of 300 bp and a minimum identity of 75%. A phylogenetic tree was constructed using the PHYLIP software (ver. 3.6) with the Kimura two-parameter model 37 and was used as an input file for the hierarchical clustering of bacterial communities in the weighted UniFrac analysis. To confirm the multiple community comparison from the UniFrac analysis, a principal coordinate analysis (PCoA) also was performed. The relative bacterial mean abundances of eyelash and tear samples from patients with and without blepharitis, shown in Figure 1, were calculated using the mean values of relative phylotypic compositions of respective eyelash and tear samples. The correlations between relative abundance of microbial communities and ocular sample type, shown in Figure 2B, were evaluated statistically by ordination biplot of redundancy analysis (RDA) with the Matlab program (ver. 6.5; MathWorks, Inc., Natick, MA). Genera unclassified by the RDP classifier and H3-T-R were not used in the analysis. 
Figure 1. 
 
Rarefaction analysis of bacterial 16S rRNA gene sequences from eyelash (A) and tear (B) samples of blepharitis patients and healthy controls. OTUs were calculated by the RDP pipeline with a 97% sequence similarity cut-off value. B, blepharitis subjects; H, healthy subjects; E, eyelash; T, tear; L, left eye; R, right eye.
Figure 1. 
 
Rarefaction analysis of bacterial 16S rRNA gene sequences from eyelash (A) and tear (B) samples of blepharitis patients and healthy controls. OTUs were calculated by the RDP pipeline with a 97% sequence similarity cut-off value. B, blepharitis subjects; H, healthy subjects; E, eyelash; T, tear; L, left eye; R, right eye.
Figure 2. 
 
Relative bacterial compositions of eyelash and tear samples from blepharitis patients and healthy controls. Partial 16S rRNA gene sequences were classified into phylum (A) and genus (B) levels using the RDP naive Bayesian rRNA Classifier based on the RDP 16S rRNA gene database at an 80% confidence threshold. Others in panel (B) are composed of the genera, each showing a percentage of reads <3.0% of the total reads in all of the subjects.
Figure 2. 
 
Relative bacterial compositions of eyelash and tear samples from blepharitis patients and healthy controls. Partial 16S rRNA gene sequences were classified into phylum (A) and genus (B) levels using the RDP naive Bayesian rRNA Classifier based on the RDP 16S rRNA gene database at an 80% confidence threshold. Others in panel (B) are composed of the genera, each showing a percentage of reads <3.0% of the total reads in all of the subjects.
Nucleotide Sequence Accession Numbers
The pyrosequencing data of the 16S rRNA genes are available publicly in the NCBI Short Read Archive (http://www.ncbi.nlm.nih.gov/sra/) under accession No. SRA050907. 
Results
Sampling and Sequencing Analysis of 16S rRNA Genes
To analyze the ocular microbial communities of humans with and without blepharitis, 22 eyelash and 22 tear samples were collected from the left and right eyes of 11 participants (seven blepharitis patients and four healthy controls), respectively (Table 1). Among a total of 44 subjects, seven tear samples did not produce sufficient 16S rRNA gene amplicons for pyrosequencing analysis. From the pyrosequencing of 37 successful PCR amplicons, a total of 96,151 sequencing reads was generated. After the removal of low quality or nonbacterial 16S rRNA sequencing reads, 79,085 high quality reads (82.25% of the total reads) with an average sequence length of approximately 486 bp and an average of >2137 reads for each sample were used for further analysis (Table 2). A rarefaction analysis using the culled 16S rRNA gene sequences was performed to assess the number of microbial communities in eyelash and tear samples were recovered from the pyrosequencing analysis (Fig. 3). Surprisingly, individual rarefaction curves of eyelash and tear samples demonstrated failures to approach asymptotes, which suggested that eyelash and tear samples had highly diverse bacterial communities, and that many unexploited OTUs still remained in the samples. The number of estimated OTUs in each subject by Chao1 richness estimator also was significantly higher than the number of observed OTUs (corresponding to 41.2%–90.7% of the estimated richness), indicating that more sequencing efforts may be required to obtain additional microbial community information. Although the number of OTUs estimated in a subject was a function of the number of pyrosequencing reads obtained, interestingly, Chao1 richness analysis demonstrated that the eyelash microbial communities of blepharitis subjects were relatively less diverse than those of healthy subjects, while the tear microbial communities of blepharitis subjects were relatively more diverse than those of healthy subjects (Table 2), which was supported more clearly by the rarefaction analysis (Fig. 3). 
Figure 3. 
 
Hierarchical clustering of bacterial communities using the weighted UniFrac UPGMA clustering method in eyelash and tear samples of blepharitis patients and healthy controls. Scale bar represents the weighted UniFrac distance. Respective symbols represent bacterial communities of eyelash (▪) and tear (•) samples from blepharitis patients, and eyelash (□) and tear (○) samples from healthy controls.
Figure 3. 
 
Hierarchical clustering of bacterial communities using the weighted UniFrac UPGMA clustering method in eyelash and tear samples of blepharitis patients and healthy controls. Scale bar represents the weighted UniFrac distance. Respective symbols represent bacterial communities of eyelash (▪) and tear (•) samples from blepharitis patients, and eyelash (□) and tear (○) samples from healthy controls.
Table 2. 
 
Summary of the Pyrosequencing and Statistical Data of Bacterial Communities of Eyelash and Tear Samples from Blepharitis Patients and Healthy Controls
Table 2. 
 
Summary of the Pyrosequencing and Statistical Data of Bacterial Communities of Eyelash and Tear Samples from Blepharitis Patients and Healthy Controls
Subject* No. of Reads No. of High Quality Reads Average Read Length (bp) OTUs† Shannon-Weaver Index (H')† Chao1† Evenness (E)†
B1-E-L 4784 4488 482 102 2.47 168.23 0.53
B1-E-R 2109 2010 477 49 1.86 70.11 0.48
B1-T-L 2161 1706 484 127 3.11 183.89 0.64
B1-T-R 2785 2454 487 140 3.07 179.06 0.62
B2-E-L 7147 6532 485 74 1.78 101.00 0.41
B2-E-R 4684 4357 495 94 1.76 123.06 0.39
B2-T-L 2519 2292 487 146 2.98 235.44 0.60
B3-E-L 4899 4608 471 103 2.06 140.00 0.45
B3-E-R 3300 3134 485 52 1.05 69.00 0.27
B4-E-L 5905 5213 507 34 0.78 37.50 0.22
B4-E-R 3360 2988 506 29 0.94 34.14 0.28
B4-T-L 1470 1240 483 73 2.57 100.00 0.60
B4-T-R 737 420 487 50 2.42 96.43 0.62
B5-E-L 1388 1312 490 56 2.82 87.63 0.70
B5-E-R 2283 2056 490 36 0.99 64.50 0.28
B5-T-L 1120 758 483 61 2.30 82.00 0.56
B5-T-R 1270 1061 487 89 2.75 109.71 0.61
B6-E-L 1453 1352 486 42 1.90 62.00 0.51
B6-E-R 2102 1942 490 88 2.19 145.27 0.49
B6-T-L 1000 857 485 98 2.89 137.00 0.63
B6-T-R 1702 1288 484 111 3.13 209.00 0.66
B7-E-L 2149 2001 485 74 2.40 111.50 0.56
B7-E-R 4051 3772 482 126 2.50 171.56 0.52
B7-T-L 1725 1606 482 177 3.60 242.21 0.70
B7-T-R 1995 1516 486 119 3.11 136.22 0.65
H1-E-L 2579 2360 496 59 2.13 78.13 0.52
H1-E-R 4726 4424 493 80 2.06 107.00 0.47
H2-E-L 2649 2358 477 128 2.29 195.50 0.47
H2-E-R 2835 2613 488 100 2.06 139.06 0.45
H2-T-R 3284 184 476 27 1.78 40.00 0.54
H3-E-L 1494 1217 480 67 2.56 92.30 0.61
H3-E-R 2006 1710 480 102 2.50 159.50 0.54
H3-T-L 1183 468 487 66 2.59 106.62 0.62
H3-T-R 1991 665 487 64 2.53 126.33 0.61
H4-E-L 1058 879 490 98 2.98 176.40 0.65
H4-E-R 1298 1100 490 120 3.08 246.00 0.64
H4-T-R 2950 144 481 28 2.44 68.00 0.73
Ocular Microbial Communities of Blepharitis and Healthy Subjects
To compare the ocular bacterial taxa compositions of blepharitis and healthy subjects, the bacterial 16S rRNA sequencing reads of individual subjects were classified using the RDP naive Bayesian rRNA Classifier at both phylum and genus levels (Fig. 4). At a 80% confidence threshold in the RDP Classifier, the 16S rRNA gene sequencing reads of eyelash and tear samples were classified into 12 bacterial phyla, and most sequences were affiliated predominantly with five phyla: Actinobacteria (0.06%–95.53%), Proteobacteria (0.45%–99.67%), Firmicutes (0.17%–84.23%), Cyanobacteria (0–44.38%), or Bacteroidetes (0–32.48%), which together accounted for 88.89% to 100% of all sequencing reads (Fig. 4A). The relative abundances of the five prevalent phyla in each subject were significantly variable, depending on individual and sample type. Interestingly, some subjects were predominated by a single phylum. For example, subject B4-E-L was predominated by Proteobacteria, with 99.67% abundance of total sequencing reads, while subject B3-E-L was predominated by Actinobacteria, with 95.53% abundance, and Proteobacteria accounting for only 3.03% of the total sequencing reads in the same sample. Eyelash samples had higher variability than tear samples in terms of the relative abundances of the prevalent phyla in each sample, possibly because eyelids have less consistent conditions due to environmental exposures, as compared to the bulbar conjunctiva. The bacterial reads belonging to Fusobacteria (0–2.61%), Planctomycetes (0–0.61%), Acidobacteria (0–1.02%), OP10 (0–0.69%), TM7 (0–0.25%), Deinococcus-Thermus (0–0.45%), and Spirochaetes (0–0.44%) also were found as minor groups. 
Figure 4. 
 
PCoA results showing the relationships of bacterial communities in eyelash and tear samples of blepharitis patients and healthy controls. The PCoA plot was constructed using the weighted UniFrac method. Respective symbols represent bacterial communities of eyelash (▪) and tear (•) samples from subjects with blepharitis, and eyelash (□) and tear (○) samples from healthy subjects.
Figure 4. 
 
PCoA results showing the relationships of bacterial communities in eyelash and tear samples of blepharitis patients and healthy controls. The PCoA plot was constructed using the weighted UniFrac method. Respective symbols represent bacterial communities of eyelash (▪) and tear (•) samples from subjects with blepharitis, and eyelash (□) and tear (○) samples from healthy subjects.
At the genus level, most 16S rRNA gene sequencing reads from eyelash and tear samples were categorized into 24 bacterial genera (Fig. 4A). Among these, five genera, Propionibacterium, Staphylococcus, Streptophyta, Corynebacterium, and Enhydrobacter, were identified as common ocular bacteria in most subjects; however, B4-E-L (0.29%) and B4-E-R (13.42%) had very low overall abundances of the five major genera because these subjects contained high proportions of the previously unclassified genera. The relative abundances of the prevalent genera and unclassified bacterial phylotypes also varied significantly depending on individual and sample type. For example, subject B4-E-L was predominated by only a single genus, Propionibacterium, with 94.48% abundance of total reads, while trace sequencing reads belonging to Staphylococcus and Corynebacterium were detected from the same subject (<0.017%). Also, the genus Propionibacterium accounted for only 0.29% in subject B5-E-R, but Staphylococcus and Corynebacterium represented 52.24% and 45.87% of the total reads in the same subject, respectively. 
Statistical Comparisons of Ocular Microbial Communities of Blepharitis and Healthy Subjects
The bacterial compositions of eyelash and tear subjects from humans with and without blepharitis were assessed statistically using a phylogeny-based metric, UniFrac based on representative sequences derived from all culled 16S rRNA gene sequences of the individual subjects. As shown in Figure 5, although there were some exceptions, intrapersonal bacterial communities of eyelash subjects were relatively well clustered compared to the interpersonal bacterial communities, which was consistent with previous results that interpersonal variability was high, whereas individuals exhibited less internal variability. 22 Bacterial communities of tear samples from blepharitis subjects were well clustered from those of eyelash samples or those of healthy tear subjects. PCoA also demonstrated that tear samples from blepharitis subjects were distinguished clearly from those of other subjects, which suggested that some genera or microbiota representing human blepharitis may be present in the conjunctiva in these patients. However, eyelash samples from subjects with blepharitis were not clearly statistically different (Fig. 6), which might be explained by the fact that microbial communities of eyelashes can be influenced easily by external factors. 
Figure 5. 
 
Relative bacterial mean abundances in eyelash and tear samples from blepharitis patients and healthy controls at phylum (A) and genus (B) levels. The relative bacterial mean abundances were calculated by the mean values of relative phylotypic compositions of respective eyelash and tear samples. Others in panel (B) are composed of the genera each showing a percentage of reads <3.0% of the total reads in all subjects.
Figure 5. 
 
Relative bacterial mean abundances in eyelash and tear samples from blepharitis patients and healthy controls at phylum (A) and genus (B) levels. The relative bacterial mean abundances were calculated by the mean values of relative phylotypic compositions of respective eyelash and tear samples. Others in panel (B) are composed of the genera each showing a percentage of reads <3.0% of the total reads in all subjects.
Figure 6. 
 
Ordination biplot of RDA showing correlations between subject types and microbial communities of Figure 5B. Subject types are represented by circles and rectangles (closed or open), and genera are represented by inverted triangles. Genera unclassified by the RDP classifier were not used in the analysis. Arrow: directions point toward maximal abundance, and their lengths are proportional to the maximal rate of change between subject types.
Figure 6. 
 
Ordination biplot of RDA showing correlations between subject types and microbial communities of Figure 5B. Subject types are represented by circles and rectangles (closed or open), and genera are represented by inverted triangles. Genera unclassified by the RDP classifier were not used in the analysis. Arrow: directions point toward maximal abundance, and their lengths are proportional to the maximal rate of change between subject types.
Although the bacterial communities of tear samples from blepharitis patients were relatively well distinguished from those of other tear and eyelash samples, their bacterial communities had too much interpersonal variability to compare their bacterial differences. Therefore, relative bacterial mean abundances of eyelash and tear samples of blepharitis patients and healthy controls were calculated (Fig. 1). At the phylum level, the results clearly showed that five phyla, Actinobacteria, Proteobacteria, Fimicutes, Cyanobacteria, and Bacteroidetes, were predominant in all eyelash and tear samples (Figs. 1A, 4A). The relative proportions of Actinobacteria in eyelash and tear samples with blepharitis were slightly lower than those in healthy subjects. The relative proportions of Proteobacteria and Fimicutes in eyelash samples from blepharitis patients and healthy controls were similar, while the relative proportions of Proteobacteria and Fimicutes in tear samples from patients with blepharitis were clearly higher than those in tear samples from healthy controls. Surprisingly, the relative proportions of Cyanobacteria, whose source may be plant material, such as pollen, in eyelash and tear samples from blepharitis patients were clearly higher than in those from healthy controls. Interestingly, tear samples of healthy controls had a significantly higher proportion of Bacteroidetes than did other tear and eyelash samples, which suggests that Bacteroidetes might be important as a resident commensal microbiota and may contribute to the prevention of blepharitis. 38 The genus level analysis demonstrated that Propionibacterium, Staphylococcus, Streptophyta, Corynebacterium, and Enhydrobacter were common ocular bacteria in all eyelash and tear samples, regardless of the presence of blepharitis (Fig. 1B); these results differed slightly from prior results that Pseudomonas, Bradyrhizobium, Propionibacterium, Acinetobacter, Corynebacterium, and Staphylococcus were dominant in healthy human conjunctiva. 24 This discrepancy may be caused by differences of individuals, sampling methods, and sample types. 
Subjects with blepharitis had lower proportions of Propionibacterium than those of healthy subject, whereas the relative proportions of Streptophyta, Corynebacterium, and Enhydrobacter in eyelash and tear samples from blepharitis patients were higher than those of healthy controls. The relative proportions of Staphylococcus in eyelash samples from blepharitis patients and healthy controls were similar, while the relative proportions of Staphylococcus in tear samples from blepharitis patients clearly were higher than those in tear samples from healthy controls. Figure 1B shows that Chryseobacterium was identified as one of the major populations from tear samples of healthy controls (6.35%). However, because its high proportion in healthy tear samples was found in only one tear sample (H3-T-R; Fig. 4B), the high proportion of Chryseobacterium was not considered to be normal flora of the ocular microbiome, and Chryseobacterium was excluded from the following discussion. The correlation between sample type and microbial community was confirmed by redundancy analysis (Fig. 2). The distributions of subject type and microbial community in the ordination space, as determined by RDA, clearly highlighted that subjects with blepharitis had more abundant Streptophyta, Corynebacterium, and Enhydrobacter than healthy subjects (Fig. 1B). The RDA results also highlighted the uniqueness of the tear samples, mainly due to the abundance of the genus Propionibacterium (Fig. 2). 
Discussion
An understanding of the ocular microbial community is essential for the prevention and treatment of blepharitis, as the ocular microbiota contributes to infection and prevention of eye diseases. Many researchers have analyzed ocular microbial communities, and members of the genera Propionibacterium, Staphylococcus, Acinetobacter, and Corynebacterium have been identified as major microbiota from eye conjunctivas or lids with blepharitis. 16,19,39,40 Comparative community analysis using culture-based approaches found that the ocular microbial communities of patients with and without blepharitis could differ in terms of their relative abundance proportions, 15,16,18 which suggested that the differences in ocular microbial communities, especially in the relative abundance of Propionibacterium and Staphylococcus, might contribute to the occurrence of blepharitis. However, culture-based approaches have many limitations in terms of the culturability of microorganisms. Recently, a culture-independent approach based on pyrosequencing demonstrated that Pseudomonas, Propionibacterium, Bradyrhizobium, Corynebacterium, Acinetobacter, Brevundimonas, and Staphylococcus, as well as many other previously undescribed bacteria, were identified from healthy human conjunctiva. 24 Therefore, in our current study, we applied a massively parallel pyrosequencing strategy to compare the ocular microbial communities of humans with and without blepharitis, which potentially will be very helpful in understanding the occurrence and treatment of human blepharitis on the basis of ocular microbial flora. 
Groden et al. 16 demonstrated that members of Propionibacterium, Corynebacterium, Staphylococcus, and Acinetobacter were identified as the most common isolates from all lids, but that normal skin bacteria, such as Staphylococcus and Propionibacterium, were isolated in greater quantities from lids with blepharitis. Kulaçoğlu et al. reported that mixed skin microbial flora, including P. acnes , were found in blepharitis patients and healthy controls, but Staphylococcus and some other skin microbial flora, such as Prevotella and Bacteriodes, were not found in healthy controls. 18 These previous reports suggest that skin microbial flora can be a cause of human blepharitis on the basis of ocular microbial flora. Our analysis demonstrated that Propionibacterium and Staphylococcus, as well as Streptophyta, Corynebacterium, and Enhydrobacter, were identified as the most common ocular microbial flora, regardless of the occurrence of blepharitis, but that their compositions were different depending on sample types and the occurrence of blepharitis (Fig. 1B). The relative proportions of Staphylococcus, Streptophyta, Corynebacterium, and Enhydrobacter were higher in subjects with blepharitis than in healthy subjects, especially in tear samples; however, surprisingly, the proportions of Propionibacterium clearly were lower in subjects with blepharitis than in healthy subjects (Fig. 1B), which suggests that Propionibacterium might be important as a resident commensal microbiota for the prevention of blepharitis. The relative proportion of Staphylococcus was clearly higher especially in tear samples from subjects with blepharitis than in healthy tear samples, which supports previous results that found that elevated levels of skin microbial flora, such as Staphylococcus, in the eye can be a cause of human blepharitis. 16,25,41 The phylotypic and statistical redundancy analyses demonstrated clearly that the relative abundances of Streptophyta, Corynebacterium, and Enhydrobacter were higher in tear samples from subjects with blepharitis than in healthy tear subjects (Figs. 1, 2), which is supported by a previous report that Corynebacterium elicited human blepharitis by immunoreactivity. 17 These results suggested that human blepharitis might be induced by infestations of mixed skin microbial flora, as well as plant pollens, dusts, and soil particles, because pollens, dusts, and soil are the main sources of the genera Streptophyta, Corynebacterium, and Enhydrobacter. In previous studies, Pseudomonas aeruginosa was cultured from blepharitis subjects 18,42 and Pseudomonas represented one of the major genera in healthy conjunctiva; however, our analysis showed that Pseudomonas was detected in minor abundance in all subjects, regardless of the occurrence of blepharitis (Figs. 4B, 1B). 
Demodex mites are the most common permanent ectoparasites in human skin. 43 They are easily found, especially from infundibular portions of pilous follicles of the eyelash, small hair sebaceous glands, meibomian glands, face, and external otic tract, where active sebum excretion provides a favorable habitat for breeding. 44 Some prior studies have reported that Demodex or Demodex-related Bacillus might contribute to the occurrence of blepharitis. 811 However, in our analysis, the incidence of Demodex mites did not demonstrate a clear correlation with ocular microbial community, although sample sizes were too small to allow for a statistical comparison. Despite the finding that Demodex mites are found more frequently in blepharitis patients, there is controversy as to whether Demodex is a cause of blepharitis, since blepharitis symptoms often are found in humans not associated with blepharitis. 12,13 Therefore, the frequent discovery of Demodex mites from blepharitis patients may not reflect the cause of blepharitis occurrence, but rather a result of blepharitis because active sebum excretion caused by blepharitis can provide favorable conditions for Demodex mites. Although Demodex mites may worsen blepharitis symptoms, additional investigations are required to clarify these hypotheses. In our current study, we compared ocular microbial communities of humans with and without blepharitis using pyrosequencing and suggested that some ocular microbiota can contribute to infection and prevention of human blepharitis. Our analysis demonstrated that many bacteria known as ocular surface pathogens were identified with high abundance in ocular samples, regardless of the occurrence of blepharitis, as reported previously, 24,26 but that their compositions were different depending on the occurrence of blepharitis. These results suggested that the balance or the commensal growth between ocular microbiota might be important for the prevention of blepharitis, because ocular health and blepharitis may depend on the interplay between the eye and the ocular microbial community. However, further studies at species or strain levels will be required to test the validity of this hypothesis. Continued investigations of ocular microbial communities are required to add valuable information for the prevention and treatment of human blepharitis, because the roles of the ocular microbial community in humans with and without blepharitis are unknown. 
Supplementary Materials
References
Rubin M Rao SN. Efficacy of topical cyclosporin 0.05% in the treatment of posterior blepharitis. J Ocul Pharmacol Ther . 2006;22:47–53. [CrossRef] [PubMed]
Elston DM. Demodex mites: facts and controversies. Clin Dermatol . 2010;28:502–504. [CrossRef] [PubMed]
Forton F Seys B. Density of Demodex folliculorum in rosacea: a case-control study using standardized skin-surface biopsy. Br J Dermatol . 1993;128:650–659. [CrossRef] [PubMed]
Lee SH Chun YS Kim JH Kim ES Kim JC. The relationship between Demodex and ocular discomfort. Invest Ophthalmol Vis Sci . 2010;51:2906–2911. [CrossRef] [PubMed]
Liu J Sheha H Tseng SC. Pathogenic role of Demodex mites in blepharitis. Curr Opin Allergy Clin Immunol . 2010;10:505–510. [CrossRef] [PubMed]
McCulley JP Shine WE. Changing concepts in the diagnosis and management of blepharitis. Cornea . 2000;19:650–858. [CrossRef] [PubMed]
Rebora A. The management of rosacea. Am J Clin Dermatol . 2002;3:489–496. [CrossRef] [PubMed]
Lacey N Delaney S Kavanagh K Powel FC. Mite-related bacterial antigens stimulate inflammatory cells in rosacea. Brit J Dermatol . 2007;157:474–481. [CrossRef]
Li J O'Reilly N Sheha H Correlation between ocular Demodex infestation and serum immunoreactivity to Bacillus proteins in patients with facial rosacea. Ophthalmology . 2010;117:870–877. [CrossRef] [PubMed]
Szkaradkiewicz A Chudzicka-Strugała I Karpiński TM Bacillus oleronius and Demodex mite infestation in patients with chronic blepharitis [published online ahead of print October 21, 2011]. Clin Microbiol Infect . doi:10.1111/j.1469-0691.2011.03704.x .
Wolf T Ophir J Avigad J Lengy J Krakowski A. The hair follicle mites (Demodex spp.). Could they be vectors of pathogenic microorganisms? Acta Derm Venereol . 1988;68:535–537. [PubMed]
Hay R. Demodex and skin infection: fact or fiction. Curr Opin Infect Dis . 2010;23:103–105. [CrossRef] [PubMed]
Kim JT Lee SH Chun YS Kim JC. Tear cytokines and chemokines in patients with Demodex blepharitis. Cytokine . 2011;53:94–99. [CrossRef] [PubMed]
Norn MS. Incidence of Demodex folliculorum on skin of lids nose. Acta Ophthamol (Copenh) . 1982;60:585–593.
Dougherty JM McCulley JP. Comparative bacteriology of chronic blepharitis. Br J Ophthalmol . 1984;68:524–528. [CrossRef] [PubMed]
Groden LR Murphy B Rodnite J Genvert GI. Lid flora in blepharitis. Cornea . 1991;10:50–53. [CrossRef] [PubMed]
Izumi K Hatano H Ito N Mizuki N. A case of Corynebacterium blepharitis resulting from long-term local immunosuppressive therapy. Folia Ophthalmol Japonica . 2006;57:205–208.
Kulaçoğlu DN Özbek A Uslu H Comparative lid flora in anterior blepharitis. Turk J Med Sci . 2001;31:359–363.
Ta CN Shine WE McCulley JP Pandya A Trattler W Norbury JW. Effects of minocycline on the ocular flora of patients with acne rosacea or seborrheic blepharitis. Cornea . 2003;22:545–548. [CrossRef] [PubMed]
Petti CA Polage CR Schreckenberger P. The role of 16S rRNA gene sequencing in identification of microorganisms misidentified by conventional methods. J Clin Microbiol . 2005;43:6123–6125. [CrossRef] [PubMed]
Tuttle MS Mostow E Mukherjee P Characterization of bacterial communities in venous insufficiency wounds by use of conventional culture and molecular diagnostic methods. J Clin Microbiol . 2011;49:3812–3819. [CrossRef] [PubMed]
Costello EK Lauber CL Hamady M Fierer N Gordon JI Knight R. Bacterial community variation in human body habitats across space and time. Science . 2009;326:1694–1697. [CrossRef] [PubMed]
Fierer N Hamady M Lauber CL Knight R. The influence of sex, handedness, and washing on the diversity of hand surface bacteria. Proc Nat Acad Sci U S A . 2008;105:17994–17999. [CrossRef]
Dong Q Brulc JM Iovieno A Diversity of bacteria at healthy human conjunctiva. Invest Ophthalmol Vis Sci . 2011;52:5408–5413. [CrossRef] [PubMed]
Graham JE Moore JE Jiru X Ocular pathogen or commensal: a PCR-based study of surface bacterial flora in normal and dry eyes. Invest Ophthalmol Vis Sci . 2007;48:5616–5623. [CrossRef] [PubMed]
Shabereiter-Gurtner C Maca S Rölleke S 16S rDNA-based identification of bacteria from conjunctival swabs by PCR and DGGE fingerprinting. Invest Ophthalmol Vis Sci . 2001;42:1164–1171. [PubMed]
Jung JY Lee SH Lee HJ Seo HY Park WS Jeon CO. Effects of Leuconostoc mesenteroides starter cultures on microbial communities and metabolites during kimchi fermentation. Int J Food Microbiol . 2012;153:378–387. [CrossRef] [PubMed]
Roesch LF Fulthorpe RR Riva A Pyrosequencing enumerates and contrasts soil microbial diversity. ISME J . 2007;1:283–290. [PubMed]
Cole JR Wang Q Cardenas E The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res . 2009;37:D141–D145. [CrossRef] [PubMed]
Wang Q Garrity GM Tiedje JM Cole JR. Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl Environ Microbiol . 2007;73:5264–5267.
Shannon CE Weaver W. The Mathematical Theory of Communication . Urbana, IL: University of Illinois Press; 1963.
Chao A. Estimating the population size for capture-recapture data with unequal catchability. Biometrics . 1987;43:783–791. [CrossRef] [PubMed]
Lozupone C Knight R. UniFrac: a new phylogenetic method for comparing microbial communities. Appl Environ Microbiol . 2005;71:8228–8235. [CrossRef] [PubMed]
Li W Godzik A. Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics . 2006;22:1658–1659. [CrossRef] [PubMed]
DeSantis TZ Hugenholtz P Keller K NAST: a multiple sequence alignment server for comparative analysis of 16S rRNA genes. Nucleic Acids Res . 2006;34:W394–W339. [CrossRef] [PubMed]
DeSantis TZ Hugenholtz P Larsen N Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Appl Environ Microbiol . 2006;72:5069–5072. [CrossRef] [PubMed]
Felsenstein J. PHYLIP (Phylogeny Inference Package), Version 3.6a . Seattle, WA: Department of Genetics, University of Washington; 2002.
Cogen AL Nizet V Gallo RL. Skin microbiota: a source of disease or defense? Br J Dermatol . 2008;158:442–255. [CrossRef] [PubMed]
McCulley JP Dougherty JM. Bacterial aspects of chronic blepharitis. Trans Ophthalmol Soc U K . 1986;105:314–318. [PubMed]
Suzuki T Sano Y Sasaki O Kinoshita S. Ocular surface inflammation induced by Propionibacterium acnes . Cornea . 2002;21:812–817. [CrossRef] [PubMed]
Karimian F Zarei-Ghanavati S A BR Jadidi K Lotfi-Kian A. Microbiological evaluation of chronic blepharitis among Iranian veterans exposed to mustard gas: a case-controlled study. Cornea . 2011;60:620–623. [CrossRef]
Giagounidis AA Meckenstock G Flacke S Pseudomonas aeruginosa blepharoconjunctivitis during cytoreductive chemotherapy in a woman with acute lymphocytic leukemia. Ann Hematol . 1997;75:121–123. [CrossRef] [PubMed]
Basta-Juzbasić A Subić JS Ljubojević S. Demodex folliculorum in development of dermatitis rosaceiformis steroidica and rosacea-related disease. Clin Dermatol . 2002;20:135–140. [CrossRef] [PubMed]
Holzchuh FG Hida RY Moscovici BK Clinical treatment of ocular Demodex folliculorum by systemic ivermectin. Am J Ophthalmol . 2011;151:1030–1034. [CrossRef] [PubMed]
Footnotes
 Supported by the Technology Development Program for Agriculture and Forestry (TDPAF) of the Ministry for Agriculture, Forestry and Fisheries, the Next-Generation BioGreen 21 Program (No. SSAC2011-PJ008220), Rural Development Administration, and the National Research Foundation (program #2011-0016922), Republic of Korea. DHO and JCK were supported by the National Research Foundation of Korea (program #2011-0016922).
Footnotes
 Disclosure: S.H. Lee, None; D.H. Oh, None; J.Y. Jung, None; J.C. Kim, None; C.O. Jeon, None
Figure 1. 
 
Rarefaction analysis of bacterial 16S rRNA gene sequences from eyelash (A) and tear (B) samples of blepharitis patients and healthy controls. OTUs were calculated by the RDP pipeline with a 97% sequence similarity cut-off value. B, blepharitis subjects; H, healthy subjects; E, eyelash; T, tear; L, left eye; R, right eye.
Figure 1. 
 
Rarefaction analysis of bacterial 16S rRNA gene sequences from eyelash (A) and tear (B) samples of blepharitis patients and healthy controls. OTUs were calculated by the RDP pipeline with a 97% sequence similarity cut-off value. B, blepharitis subjects; H, healthy subjects; E, eyelash; T, tear; L, left eye; R, right eye.
Figure 2. 
 
Relative bacterial compositions of eyelash and tear samples from blepharitis patients and healthy controls. Partial 16S rRNA gene sequences were classified into phylum (A) and genus (B) levels using the RDP naive Bayesian rRNA Classifier based on the RDP 16S rRNA gene database at an 80% confidence threshold. Others in panel (B) are composed of the genera, each showing a percentage of reads <3.0% of the total reads in all of the subjects.
Figure 2. 
 
Relative bacterial compositions of eyelash and tear samples from blepharitis patients and healthy controls. Partial 16S rRNA gene sequences were classified into phylum (A) and genus (B) levels using the RDP naive Bayesian rRNA Classifier based on the RDP 16S rRNA gene database at an 80% confidence threshold. Others in panel (B) are composed of the genera, each showing a percentage of reads <3.0% of the total reads in all of the subjects.
Figure 3. 
 
Hierarchical clustering of bacterial communities using the weighted UniFrac UPGMA clustering method in eyelash and tear samples of blepharitis patients and healthy controls. Scale bar represents the weighted UniFrac distance. Respective symbols represent bacterial communities of eyelash (▪) and tear (•) samples from blepharitis patients, and eyelash (□) and tear (○) samples from healthy controls.
Figure 3. 
 
Hierarchical clustering of bacterial communities using the weighted UniFrac UPGMA clustering method in eyelash and tear samples of blepharitis patients and healthy controls. Scale bar represents the weighted UniFrac distance. Respective symbols represent bacterial communities of eyelash (▪) and tear (•) samples from blepharitis patients, and eyelash (□) and tear (○) samples from healthy controls.
Figure 4. 
 
PCoA results showing the relationships of bacterial communities in eyelash and tear samples of blepharitis patients and healthy controls. The PCoA plot was constructed using the weighted UniFrac method. Respective symbols represent bacterial communities of eyelash (▪) and tear (•) samples from subjects with blepharitis, and eyelash (□) and tear (○) samples from healthy subjects.
Figure 4. 
 
PCoA results showing the relationships of bacterial communities in eyelash and tear samples of blepharitis patients and healthy controls. The PCoA plot was constructed using the weighted UniFrac method. Respective symbols represent bacterial communities of eyelash (▪) and tear (•) samples from subjects with blepharitis, and eyelash (□) and tear (○) samples from healthy subjects.
Figure 5. 
 
Relative bacterial mean abundances in eyelash and tear samples from blepharitis patients and healthy controls at phylum (A) and genus (B) levels. The relative bacterial mean abundances were calculated by the mean values of relative phylotypic compositions of respective eyelash and tear samples. Others in panel (B) are composed of the genera each showing a percentage of reads <3.0% of the total reads in all subjects.
Figure 5. 
 
Relative bacterial mean abundances in eyelash and tear samples from blepharitis patients and healthy controls at phylum (A) and genus (B) levels. The relative bacterial mean abundances were calculated by the mean values of relative phylotypic compositions of respective eyelash and tear samples. Others in panel (B) are composed of the genera each showing a percentage of reads <3.0% of the total reads in all subjects.
Figure 6. 
 
Ordination biplot of RDA showing correlations between subject types and microbial communities of Figure 5B. Subject types are represented by circles and rectangles (closed or open), and genera are represented by inverted triangles. Genera unclassified by the RDP classifier were not used in the analysis. Arrow: directions point toward maximal abundance, and their lengths are proportional to the maximal rate of change between subject types.
Figure 6. 
 
Ordination biplot of RDA showing correlations between subject types and microbial communities of Figure 5B. Subject types are represented by circles and rectangles (closed or open), and genera are represented by inverted triangles. Genera unclassified by the RDP classifier were not used in the analysis. Arrow: directions point toward maximal abundance, and their lengths are proportional to the maximal rate of change between subject types.
Table 1. 
 
Demographics of Blepharitis Patients and Healthy Controls for Eyelash and Tear Sampling
Table 1. 
 
Demographics of Blepharitis Patients and Healthy Controls for Eyelash and Tear Sampling
Group Subject No. Age (y) Sex Demodex* Allergy†
Left Right
Blepharitis patients (B) 1 63 Female +
2 66 Male + +
3 65 Male + + +
4 66 Male + + +
5 69 Male + +
6 64 Male + +
7 76 Female
Healthy controls (H) 1 54 Male
2 76 Female
3 25 Male
4 27 Male +
Table 2. 
 
Summary of the Pyrosequencing and Statistical Data of Bacterial Communities of Eyelash and Tear Samples from Blepharitis Patients and Healthy Controls
Table 2. 
 
Summary of the Pyrosequencing and Statistical Data of Bacterial Communities of Eyelash and Tear Samples from Blepharitis Patients and Healthy Controls
Subject* No. of Reads No. of High Quality Reads Average Read Length (bp) OTUs† Shannon-Weaver Index (H')† Chao1† Evenness (E)†
B1-E-L 4784 4488 482 102 2.47 168.23 0.53
B1-E-R 2109 2010 477 49 1.86 70.11 0.48
B1-T-L 2161 1706 484 127 3.11 183.89 0.64
B1-T-R 2785 2454 487 140 3.07 179.06 0.62
B2-E-L 7147 6532 485 74 1.78 101.00 0.41
B2-E-R 4684 4357 495 94 1.76 123.06 0.39
B2-T-L 2519 2292 487 146 2.98 235.44 0.60
B3-E-L 4899 4608 471 103 2.06 140.00 0.45
B3-E-R 3300 3134 485 52 1.05 69.00 0.27
B4-E-L 5905 5213 507 34 0.78 37.50 0.22
B4-E-R 3360 2988 506 29 0.94 34.14 0.28
B4-T-L 1470 1240 483 73 2.57 100.00 0.60
B4-T-R 737 420 487 50 2.42 96.43 0.62
B5-E-L 1388 1312 490 56 2.82 87.63 0.70
B5-E-R 2283 2056 490 36 0.99 64.50 0.28
B5-T-L 1120 758 483 61 2.30 82.00 0.56
B5-T-R 1270 1061 487 89 2.75 109.71 0.61
B6-E-L 1453 1352 486 42 1.90 62.00 0.51
B6-E-R 2102 1942 490 88 2.19 145.27 0.49
B6-T-L 1000 857 485 98 2.89 137.00 0.63
B6-T-R 1702 1288 484 111 3.13 209.00 0.66
B7-E-L 2149 2001 485 74 2.40 111.50 0.56
B7-E-R 4051 3772 482 126 2.50 171.56 0.52
B7-T-L 1725 1606 482 177 3.60 242.21 0.70
B7-T-R 1995 1516 486 119 3.11 136.22 0.65
H1-E-L 2579 2360 496 59 2.13 78.13 0.52
H1-E-R 4726 4424 493 80 2.06 107.00 0.47
H2-E-L 2649 2358 477 128 2.29 195.50 0.47
H2-E-R 2835 2613 488 100 2.06 139.06 0.45
H2-T-R 3284 184 476 27 1.78 40.00 0.54
H3-E-L 1494 1217 480 67 2.56 92.30 0.61
H3-E-R 2006 1710 480 102 2.50 159.50 0.54
H3-T-L 1183 468 487 66 2.59 106.62 0.62
H3-T-R 1991 665 487 64 2.53 126.33 0.61
H4-E-L 1058 879 490 98 2.98 176.40 0.65
H4-E-R 1298 1100 490 120 3.08 246.00 0.64
H4-T-R 2950 144 481 28 2.44 68.00 0.73
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×