- Top of page
- 2Materials and methods
Mycoplasma fermentans and other Mycoplasma species are colonizers of human mucosal surfaces and may be associated with human immunodeficiency virus infection. While many infectious agents have been described in different percentages of patients with Chronic Fatigue Syndrome (CFS), little is known about the prevalence of mycoplasmas and especially M. fermentans in CFS patients. A polymerase chain reaction (PCR)-based assay was used to detect Mycoplasma genus and M. fermentans genomes in peripheral blood mononuclear cells (PBMC) of CFS patients. Blood was collected from 100 patients with CFS and 50 control subjects. The amplified products of 717 bp of Mycoplasma genus, and 206 bp of M. fermentans were detected in DNA purified from blood samples in 52% and 34% of CFS samples, respectively. In contrast, these genomes were found in only 14% and 8% of healthy control subjects respectively (P<0.0001). All samples were confirmed by Southern blot with a specific probe based on internal sequences of the expected amplification product. Several samples, which were positive for Mycoplasma genus, were negative for M. fermentans indicating that other Mycoplasma species are involved. A quantitative PCR was developed to determine the number of M. fermentans genome copies present in 1 μg of DNA for controls and CFS patients. Mycoplasma copy numbers ranging from 130 to 880 and from 264 to 2400 were detected in controls and CFS positive subjects, respectively. An enzyme immunoassay was applied for the detection of antibodies against p29 surface lipoprotein of M. fermentans to determine the relationship between M. fermentans genome copy numbers and antibody levels. Individuals with high genome copy numbers exhibited higher IgG and IgM antibodies against M. fermentans specific peptides. Isolation of this organism by culture from clinical specimens is needed in order to demonstrate specificity of signal detected by PCR in this study.
- Top of page
- 2Materials and methods
Chronic Fatigue Syndrome (CFS) is an illness with increasingly reported frequency in the USA and other industrialized countries . CFS is characterized by prolonged and debilitating fatigue with multiple non-specific symptoms such as headaches, recurrent sore throats, muscle and joint pains and cognitive complaints. Profound fatigue, the hallmark of the disorder, can come on either suddenly or gradually and it persists throughout the illness. Unlike the short-term disability of an acute viral infection, for example, CFS symptoms by definition linger for at least six months and often for years . Physicians can evaluate patients with persistent fatigue of undetermined cause using guidelines developed by the international CFS study group .
Despite multidisciplinary investigations of CFS, its etiology remains unknown. Similarly, no specific diagnostic tests or therapies for CFS exist. In about one third of cases, the sudden onset follows a respiratory, gastrointestinal, or other acute infection with flu-like symptoms, including mononucleosis . No published data implicate a specific virus or other microbes as the cause of CFS. However, it appears that infectious agents, among other stressors, can precipitate the syndrome . A variety of common viruses can be reactivated in some CFS patients, including the HTLV-II, EBV, cytomegalovirus, herpes simplex viruses 1 and 2, and human herpes viruses 6, 7, and 8 [6–12]. Most investigators believe that virus reactivation could be occurring secondarily to some immunologic disturbance [5, 13].
Mycoplasma fermentans was chosen for this study of CFS for the following reasons. It colonizes the human mucosal tissues of healthy adults and under certain conditions can invade the host cells . It is a potential pathogen in humans and in non-human primates . An infectious agent isolated from a patient with AIDS  was later identified as a unique strain called M. fermentans, the (incognitus) strain . Blanchard and Montagnier  suggested that mycoplasmas may play a role as a cofactor in one of the early stages of lentiviruses' life cycle and promote the disease in HIV-infected patients. Coinfection with M. fermentans significantly enhances the ability of HIV-I to induce cytopathic effects on human T-lymphocytes in vitro [15–18]. In the presence of this organism, syncytium formation of HIV-infected T-cells was eliminated, despite prominent cell death. In addition, a factor produced by M. fermentans inhibited the standard reverse transcriptase enzyme assay, while replication and production of infectious HIV-I particles continued during the coinfection. In addition, it was reported that the cell killing effect of HIV-I in cultures could be significantly reduced by tetracycline [17–19]. The tetracycline-treated cultures continued to produce a high titer of HIV-I in the absence of any cytocidal effect . The authors concluded that the HIV-I copathogenicity in human lymphocytes and monocytic cells appeared to be a mycoplasma-dependent process. The actual mechanisms responsible for the synergistic effects of M. fermentans on HIV-I pathogenicity and lymphocyte killing are not clear. However, in a recent study based on antibiotic resistance , it was suggested that one particular strain of M. fermentans was, in fact, a tissue culture contaminant and was not derived from the patients . Moreover, M. fermentans DNA was recently reported in 8% of healthy persons, in 5.8% of AIDS patients and in 15% of patients with sexually transmitted diseases [20, 21].
This study was undertaken to determine the prevalence of Mycoplasma genus and M. fermentans genomes in two different groups of patients who were classified as having either typical (with no tender points) or atypical (with 6 or 7 tender points) CFS, and their comparison to healthy subjects using qualitative and quantitative PCR methodology. The development of this quantitative PCR assay may enable us and other investigators to study the clinical significance of M. fermentans in CFS and evaluate antimicrobial therapy.
2Materials and methods
- Top of page
- 2Materials and methods
A total of 100 CFS patients, which consisted of 56 female and 44 males, with a median age of 49 (23–65) were chosen for this study according to a case definition of CFS established by the Center for Disease Control . This clinical evaluation and classification was performed based on: (i) history and physical examination, (ii) mental status examination, and (iii) laboratory tests to exclude other diagnoses. Patients were determined to have Chronic Fatigue based on the following criteria established by the CDC.
Unexplained persistent or relapsing fatigue of new or definite onset that is not due to ongoing exertion, that is not relieved by rest, and that results in a substantial reduction in previous levels of activity.
Four or more of the following symptoms concurrently for six months or longer: (a) impaired memory or concentrations were enough to reduce levels of occupational, social, or personal activities; (b) sore throat; (c) tender cervical or axillary lymph nodes; (d) muscle pain; (e) multiple pain without joint swelling or redness; (f) new headaches; (g) unrefreshing sleep; (h) post-exertion malaise lasting more than 24 h.
Patients were classified as atypical CFS if, in addition to the above criteria, they had 6 or 7 tender points. Each patient had been ill for at least one year. Out of 100 CFS subjects, 78% were Caucasian; 12% were African-American; and 10% were Asian. Racial background did not affect the results. Control subjects (n=50) from the same geographical area, with similar age and race distribution, were chosen from individuals who went to different clinics for yearly checkups and did not have any of the above symptomatologies for chronic fatigue. Blood samples were obtained from all participants by their examining physicians in ACD containing vacutainers under sterile conditions and were sent to our laboratory for biochemical and immune function assays and for mycoplasma detection via PCR. Control specimens were drawn under similar conditions. All samples were shipped at ambient temperature and delivered within 24 h from the time the blood was drawn. While individuals performing the tests were not aware of subjects' status and handled the samples according to accession numbers, the senior author was aware of their classifications. Therefore, the study was not 100% blinded.
2.1Mycoplasma and bacterial strains
Mycoplasma fermentans ATCC 19989, M. hominis ATCC 23114 and M. orale ATCC 23714 were obtained from American Type Culture Collection, Rockville, MD, USA. DNA was prepared by phenol-chloroform extraction following a two-hour incubation at 55°C in 1 M Tris, 0.5 M EDTA (TE) buffer, pH 8.0, with 1% SDS and 50 μg ml−1 proteinase K. After precipitation in ethanol, the DNA was resuspended in TE buffer. The Mycoplasma fermentans DNA was quantified spectrophotometrically and used to prepare positive controls of known mycoplasma cell copy numbers ranging from 0 to 1000 copies μl−1. The DNA extracted from each of the Mycoplasma species served as positive controls for the Mycoplasma genus PCR and was also used to determine the specificity of the Mycoplasma fermentans PCR. In addition, human lung carcinoma cell line, Staphylococcus pasteurii, B. subtilis, Enterococcus faecalis, Streptococcus sp., Clostridium difficile, Clostridium perfringens, Clostridium ramosum, Clostridium innocuum, Rhodococcus sp., E. coli, Saccharomyces cerevisiae, Candida albicans, and Aspergillus flavus were all purchased from the American Type Culture Collection and used as controls in the PCR assay.
Blood samples were obtained in sterile 10-ml tubes containing ACD solution A (Becton-Dickinson, NJ, USA). Peripheral blood mononuclear cells (PBMC) were isolated using Histopaque (Sigma, St. Louis, MO, USA). PBMC were lysed using a lysis buffer consisting of 1 M Tris, 0.5 M EDTA, 1% SDS and 50 μg ml−1 proteinase K solution (pH 8.5). Phenol-chloroform extraction was followed by precipitation with 3 M sodium acetate and 95% ethanol. DNA was pelleted, then resuspended to a final concentration of 0.2 mg ml−1 using a Tris-EDTA buffer. This DNA extraction method yielded between 40–50 μg DNA from 10 ml of blood.
Amplification of mycoplasmal 16S rRNA sequences which can identify mycoplasmas at both the genus and species levels, were used for PCR reactions [22–25]. Computer alignment studies of these rRNA sequences have revealed the existence of regions with highly conserved sequences and regions that display sequence variability at the genus and species levels. The unique sequence features which have been extensively described allow for the selection of genus- and species-specific primers for the PCR test. For the amplification of rDNA sequences, without prior transcription of the rRNA to cDNA, the PCR assay was performed in 100 μl of reaction mixture containing 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 2.5 mM MgCl2, 0.01% gelatin, 200 μM dNTP, 2.5 units of Taq DNA polymerase and 50 pmol of each Mycoplasma genus primer 5′-ACTCCTACGGGAGGCAGCAGTA-3′, and 5′-TGCACCATCTGTCACTCTGTTAACCTC-3′, which amplify a 717-bp region from the genome of each member of the genus Mycoplasma.
To prevent non-specific annealing of the primers, the DNA was always added last, while the reaction mixture was kept at 94°C. The reaction mixtures were placed in a GENE AMP 9600 thermal cycler (Perkin Elmer; Brichburg, NJ, USA). The thermal profile involved an initial denaturation step at 94°C for 3 min followed by 40 cycles of denaturation at 94°C for 1 min, primer annealing at 60°C for 1 min and primer extension at 72°C for 2 min. The cycling was followed by a final extension step at 72°C for 10 min. To prevent contamination, a strict spatial separation of the different technical steps involved in PCR was maintained.
2.4Species-specific identification of M. fermentans
For identification of M. fermentans, each reaction was set up in a final volume of 100 μl consisting of 1 μg of sample DNA, 10 μl of 10× PCR buffer (Amersham, Arlington Heights, IL, USA), 2.5 mM MgCl2, each dNTP at 200 μm, 2.5 U Taq DNA polymerase and 50 pmol each oligonucleotide primer. Each primer (Research Genetics, Huntsville, AL, USA), 5′-GGACTATTGTCTAAACAATTTCCC-3′ and 5′-GGTTATTCGATTTCTAAATCGCCT-3′, was 24 bp long. This primer set flanks a 206-bp region in the M. fermentans genome [24, 25].
The samples were placed in the same thermal cycler and heated to 94°C for 3 min. The cycling profile consisted of 45 cycles of 94°C, 56°C and 72°C for 35 s, 45 s and 1 min, respectively. The cycling was followed by a final extension step at 72°C for 10 min. Aliquots of amplified samples (20 μl) were analyzed by electrophoresis on a 1.5% agarose gel stained with ethidium bromide.
2.5Southern blot analysis of amplified DNA
For confirmation of PCR results, Southern blot analysis was applied. The amplified PCR products were transferred to nylon filters by using 0.4 M NaOH. After UV crosslinking, filters were washed with 2× SSC (standard saline citrate). Filters were then prehybridized for 2 h at 68°C. The hybridization solution consisted of 5× SSC; blocking reagent (Boehringer Mannheim; Indianapolis, IN, USA), 0.5% (w/v); N-laurylsarcosine, Na salt, 0.1% (w/v); SDS, 0.02% (w/v). Filters were incubated for 24 h with 2.5 ml/100 cm2 hybridization solution containing 150 ng of labeled DNA probe and 450 ng of salmon sperm DNA. A 73-bp probe was generated (as described by Hawkins ) using the PCR reaction with primers (5′-GATGAGTGTATTGTCATCC-3′, 5′-AACGTAGAAGAGAATGGC-3′), which were internal to those used for the targeted DNA sequence described above. The probe was labeled with psoralen biotin using the rad-free system (Schleicher&Schuell; Keene, NH, USA). After hybridization, filters were washed with 2× SSC; SDS 0.1% (2×5 min at room temperature (RT)) followed by 2× SSC containing SDS 0.1% (2×15 min at 68°C). Hybrids were detected by using Lumi-phos 530 chemiluminescent substrate sheet. Gel electrophoresis/Southern blot reactions were positive if a band was noted on the gel and hybridization occurred on the nylon membrane .
2.6Quantitative competitive PCR
Positive samples were then subjected to a second round of amplification in order to quantitate the number of infectious agents present in 1 μg of DNA. The reaction conditions for the quantitative PCR were the same as previously described. The only change in the reaction components was the addition of a neutral DNA fragment of 348 bp flanked by the same target sequences as the mycoplasma 206-bp fragment. The control DNA was constructed as described by the manufacturers using the PCR mimic construction kit (Clontech; Palo Alto, CA, USA) and a set of composite primers which contain the target primer sequence 5′-GGACTATTGTCTAAACAATTTCCCCGCAAGTGAAATCTCCTCCG-3′ and 5′-GGTTATTCGATTTCTAAATCGCCTGGGACAAGATACTCATCTGC-3′. Once the quantity of control DNA was determined by ethidium bromide staining and spectrophotometry, serial dilutions were prepared and added to the reaction mixtures in a range of 0–1.0×104 molecules per tube. The quantity of mycoplasma DNA was determined by densitometry, construction of the standard curve and finding the molecular concentration of the unknown.
2.7Quantitative dot blot analysis
A quantitative dot blot assay was also used in order to verify the results of the quantitative competitive PCR. Known quantities of purified Mycoplasma fermentans DNA were added to human genomic DNA which was previously determined to be negative for all mycoplasma species by PCR. The mycoplasma DNA was added to negative control DNA ranging from 0 to 5000 copies per μg of human genomic DNA to serve as a quantitation control ladder. These samples were subjected to the Mycoplasma fermentans specific PCR which was previously described along with CFS and non-CFS DNA samples in duplicate (Fig. 3). The entire 100-μl volume of each PCR reaction tube was blotted on a nylon membrane using a Bio-Rad dot blot apparatus. The membrane was then probed by the same method previously described for the Southern blot. The signals from the dot blot assay following hybridization with the M. fermentans probe were analyzed by densitometry. The unknown CFS and non-CFS mycoplasma positive samples were plotted against a standard curve generated from the quantitation control ladder signals. The average densities of the duplicate signals from each sample were used to extrapolate the mycoplasma cell copy numbers.
Figure 3. Southern dot blot analyses of Mycoplasma fermentans genome copy numbers of controls and CFS patients from Table 2 with different genome copy numbers. Rows A1–A12 represent the M. fermentans quantitation ladder for calculation of genome copy number. A1=5000; A2=2500; A3=1250; A4=1000; A5=750; A6=500; A7=400; A8=300; A9=200; A10=100; A11=50; A12=0. Standard curve was generated and the calculated results for the clinical specimens were: rows C1,2=620; C3,4=310; C5,6=810; C7,8=negative; C9,10=950; D1,2=150; D3,4=negative; D5,6=230; D7,8=680; D9,10=240; D11,12=860; E1,2=840; E3,4=negative; E5,6=1720; E7,8=negative; E9,10=1550; E11,12=negative.
Download figure to PowerPoint
2.8Detection of M. fermentans-specific antibodies
Two peptides from the P29 surface lipoprotein cDNA sequence of M. fermentans PG18 or incognitus strain which originally was cultivated from an HIV-positive individual by Shiyh-Ching Lo, at the Armed Forces Institute of Pathology , were used in ELISA assay. These peptides which were chosen from two different regions of the cloned and sequenced P29 gene  were synthesized by Research Genetics (Huntsville, AL, USA). The synthesized peptides were analyzed for purity by mass spectroscopy and were 78 and 81% pure, respectively. Different microtiter wells were coated with 1 μg of each peptide and the classical indirect ELISA was applied for the measurements of IgG and IgM antibodies in control and CFS patients sera. Similar ELISA was applied to these sera by using M. fermentans ATCC 19989 lysate as a crude antigen.
The Mann-Whitney U-test was used for the generation of P values.
- Top of page
- 2Materials and methods
Leukocytes isolated from 100 patients with Chronic Fatigue Syndrome (50 patients with typical and 50 patients with atypical CFS) and 50 control subjects were used for the detection of Mycoplasma genus and Mycoplasma fermentans by PCR. The oligonucleotide primers for Mycoplasma genus and M. fermentans are highly specific since no amplification product was detected on the agarose gel when 500 ng of purified DNA from the following microorganisms Staphylococcus pasteurii, B. subtilis, Enterococcus faecalis, Streptococcus sp., Clostridium difficile, Clostridium perfringens, Clostridium ramosum, Clostridium innocuum, Rhodococcus sp., E. coli, Saccharomyces cerevisiae, Candida albicans, and Aspergillus flavus were subjected to PCR under the same conditions.
To verify the sensitivity of the PCR we added genomic DNA from M. hominis, M. orale, E. coli and human lung carcinoma cell line HTB119 to the amplification reaction mixture similar to Berg et al. . Aliquots of 1 μg of extracted DNA from M. hominis, M. orale, E. coli or human cells were mixed with serial dilutions of M. fermentans DNA, and amplification was performed as before. The sensitivity of the mycoplasma PCR was not affected under these conditions. The predicted PCR amplification product of 206 bp, reflecting the M. fermentans genome, was detected by agarose gel electrophoresis and dot blot from the reaction mixtures (data not shown). These same conditions were then applied to 11 blood samples from patients with CFS; the PCR results are shown in Fig. 1A. A 717-bp band corresponding to Mycoplasma genus DNA was detected in the PCR positive control samples, as well as CFS patients 2, 4, 5, 7, and 9. The 206 bp of M. fermentans was detected in CFS patients 4, 7, and 9, and was further confirmed by Southern blot analysis (Fig. 1 B). These results mean that sample numbers 2 and 5 showed positive for Mycoplasma genus, but not for M. fermentans (Fig. 1).
Figure 1. A: Detection of Mycoplasma genus and Mycoplasma fermentans in blood samples of patients with Chronic Fatigue Syndrome. Ethidium bromide stained products on a 1.5% agarose gel. Lane M=DNA size marker and lane C=positive controls containing 5 pg of Mycoplasma genus (717 bp) and Mycoplasma fermentans (206 bp) DNA, respectively. Lanes 1, 3, 6, 8, 10 and 11 are negative for Mycoplasma genus and Mycoplasma fermentans. Lanes 4, 7 and 9 are positive for both Mycoplasma genus and Mycoplasma fermentans. B: Southern blot analysis of Mycoplasma fermentans shown in lanes: C: control Mycoplasma fermentans DNA; 4, 7 and 9 are representative of positive samples.
Download figure to PowerPoint
This qualitative PCR for Mycoplasma genus and M. fermentans was then applied to all of the blood samples from patients and controls. Data summarized in Table 1 show that Mycoplasma genus and M. fermentans genomes were detected with much higher frequencies in the CFS groups than in the healthy control subjects. In the control group, 7 out of 50 (14%) and 4 out of 50 samples (8%) were positive by PCR for Mycoplasma genus and M. fermentans genomes, respectively. The percentage of positive mycoplasma genomes was 3.6–4-fold higher in patients with CFS, respectively (Table 1). These differences between the patients as compared to the control subjects were statistically significant (P<0.0001).
Table 1. Percentage of Mycoplasma genus and Mycoplasma fermentans in patients with chronic fatigue syndrome and control subjects
|Specimens||Number and percentage of specimens positive by PCR for|
| ||Mycoplasma genus||M. fermentans|
|Control subjects (n=50)|| 7 (14%)|| 4 (8%)|
| || || |
|Typical CFS patients (n=50)||27 (54%)||18 (36%)|
|Atypical CFS patientsa (n=50)||25 (50%)||16 (32%)|
In order to quantitate the number of infectious agents present in 1 μg of DNA, a quantitative PCR assay was developed in which the same sets of primers from M. fermentans and the control DNA were constructed by the addition of a neutral DNA fragment of 348 bp. Results depicted in Fig. 2 showed that one CFS patient had 870 copies of Mycoplasma fermentans genome per μg DNA extracted from about 2×105 white blood cells. Using a similar curve, mycoplasma cell copy numbers from 10 CFS patients and four controls who were positive for M. fermentans were quantitated. Data, depicted in Table 2, shows the genome copy numbers of M. fermentans genome in controls and CFS groups. The mean±standard deviation of M. fermentans genome copy number for the CFS group was 1052±694 and 522±315 for control subjects. Although these detected copy numbers seem to be high, presently we do not know what percentage represents living mycoplasma cells. Southern blot analyses of the PCR product shown in Table 2 were confirmed using M. fermentans positive control quantitation ladder. Mycoplasma positive samples from the same blot were compared to the quantitation ladder and results were calculated (Fig. 3). Calculated results from the dot blot quantitation shown in Fig. 3 were not significantly different from the quantitative competitive PCR results, indicating correlation between the two assays.
Figure 2. Agarose gel electrophoresis of a quantitative competitive PCR result from a mycoplasma positive CFS patient. The internal control of 348 bp is present at 5000, 2500, 1000, 500, 250 and 0 copies in lanes 1–6, respectively. The 206-bp product from the M. fermentans genome is increasingly visible in lanes 1–6. The point at which the molar concentration of the internal control and the target product are equivalent by densitometry is used to determine the mycoplasma cell copy number.
Download figure to PowerPoint
Table 2. Mycoplasma fermentans genome copy number per μg DNA and antibodies
|Sample||Control subjects||CFS patients|
| ||Genome number||IgG against (OD)||IgM against (OD)||Genome number||IgG against (OD)||IgM against (OD)|
| || ||Lysate||Peptide 1||Peptide 2||Lysate||Peptide 1||Peptide 2|| ||Lysate||Peptide 1||Peptide 2||Lysate||Peptide 1||Peptide 2|
| 1||130||0.25||0.16||0.19||0.36||0.21||0.19|| 600||0.86||0.32||0.36||0.78||0.42||0.35|
| 4||630||0.21||0.20||0.24||0.46||0.17||0.22|| 320||0.61||0.21||0.34||0.49||0.27||0.22|
| 5||N||0.37||0.17||0.19||0.42||0.16||0.18|| 910||0.57||0.33||0.29||0.57||0.35||0.30|
| 7||N||0.41||0.20||0.23||0.34||0.18||0.26|| 870||0.64||0.36||0.27||0.89||0.27||0.32|
| 9||N||0.48||0.23||0.22||0.44||0.29||0.25|| 680||0.47||0.33||0.41||0.61||0.22||0.19|
Preliminary supportive evidence for specificity of positive signal detected in our PCR assay is provided in Table 2. Using M. fermentans lysate and two specific peptides from P29 surface lipoprotein sequence, ELISA assays were developed and IgG as well as IgM antibodies were measured in sera of controls and CFS patients. Overall, IgG as well as IgM optical densities against all three antigens were higher in CFS patients than in control subjects. Moreover, three patients (3, 6, 8) in Table 2 with genome copy numbers of 2400, 1640, and 1840, respectively, showed the highest optical densities for IgG and IgM against the lysate as well as the M. fermentans specific peptides. While IgG and IgM optical densities (OD) against M. fermentans lysate in some control subjects were twice or more higher than the ELISA background (about 0.2) reading, none of the controls including the four with positive genome copy numbers had significant elevation in IgG and IgM antibodies against M. fermentans peptides. The difference between the optical densities for IgG and IgM against Mycoplasma fermentans in samples from CFS patients who were negative for M. fermentans by PCR were not statistically significant from non-CFS patients who were also PCR negative.
- Top of page
- 2Materials and methods
Numerous methods for detecting Mycoplasma fermentans infection have been developed. Direct tests are based on microbiological culture, and indirect tests are based on measurement of specific markers or characteristics associated with mollicutes, including DNA fluorochrome staining, DNA probes, enzyme-linked immunoabsorbent assay (ELISA), immunofluorescence, electron microscopy, autoradiography, and biochemical assays [28–30]. Although efforts have focused on the improvement of these techniques, accurate detection of mycoplasma, especially M. fermentans, is difficult . Recently, the application of PCR-based methods of detection has attracted much attention because of their excellent sensitivity and specificity.
Using a PCR-based assay M. fermentans was reported in 44.5% of the saliva samples collected from healthy adults . This and other studies by Katseni  suggested that saliva, and by extension the mouth and oropharynx, may be a significant reservoir of M. fermentans in the human population and that it may be a benign colonizer of the oral mucosa in the healthy young and middle-aged adult population. Therefore, laboratory personnel that are carriers of M. fermentans could be the main source of tissue culture contamination and possibly PCR false positive results. In our assay, we excluded this possibility since both controls' and CFS patients' blood was drawn simultaneously and under similar aseptic conditions using vacutainers. Moreover, in addition to the proper controls used in PCR assay the PBMC separation by Histopaque was performed in a biological safety cabinet and the DNA was extracted immediately.
M. fermentans DNA also was reported in the PBMC of 8% healthy controls, 5.8% in HIV infected patients and in 15% of patients with sexually transmitted diseases . In the present study the percent positive PCR for M. fermentans in control subjects is in complete agreement with the study by Kovacic et al. . We believe that the prevalence of this organism in HIV infected patients and sexually transmitted diseases was underestimated by these investigators since the majority of these patients were under antimicrobial treatment.
Although some believe M. fermentans to have a benign commensal relationship with the human host, other reports describe the pathogenic potential of M. fermentans[32–37]. In few studies and case reports M. fermentans was found to be the only infectious agent identified in blood samples or tissue biopsies of previously healthy non-AIDS patients who suffered from chronic fatigue or flu-like illness [13, 33, 34]. These patients recovered from this illness after adequate antimicrobial therapy [33, 34]. However, for demonstrating the cause and effects relationship these findings have not been adequately confirmed. In this study, we tried to determine the prevalence of mycoplasma in patients with classical CFS (based on CDC criteria) and CFS patients which in addition to CDC classification had 6 or 7 tender points (atypical CFS) and were compared to the matched controls. According to the American College of Rheumatology criteria for the diagnosis of fibromyalgia , findings of 11 or more out of 18 designated tender point areas should be evident in order to be classified as fibromyalgia. Since in this group of CFS patients only 6 or 7 tender points were observed we could not classify them as CFS with fibromyalgia, therefore, the term atypical CFS was used. While the number and percentage of specimens positive by PCR for Mycoplasma genus and M. fermentans (Table 1) were not significantly different in both CFS groups, the overall percent positive in CFS groups was 3–4-fold higher than control (P<0.0001). This similarity in percent positive mycoplasma in both CFS groups indicates lack of involvement of this organism in the finding of 6 or 7 tender points described in the atypical CFS group. In addition, phylogenetically closely related and non-related microorganisms were studied for the exclusions of possible cross reaction of the primers used in this PCR. M. fermentans primers used in our assay did not form products with other mycoplasmas including M. hominis and M. orale (data not shown), indicating specificity of the PCR assay. We did not observe any PCR result that was positive for M. fermentans genome but negative for Mycoplasma genus. On the contrary, we had many cases that were positive for Mycoplasma genus but not for M. fermentans, indicating that other mycoplasma species are present in PBMC of patients with chronic fatigue. Furthermore, similar to Hawkins et al. , we were able to confirm the results by Southern blot with the use of an internal probe. Additional work is in progress to define the species of mycoplasmas other than M. fermentans genome in CFS patients. We do not know why a small percentage of healthy individuals are carrying the genome of mycoplasma in their blood and whether or not this will progress to a disease. Also, presently we do not know what percentage of this detected signal represents living mycoplasma cells since several attempts failed to isolate this organism by classical mycoplasma culture techniques. To further study the clinical significance of M. fermentans in CFS and to evaluate antimicrobial therapy, we developed a quantitative competitive PCR for accurate detection of this organism (Fig. 2). Using this PCR, the genome copy number of M. fermentans was determined for PCR positive samples and was confirmed by a quantitative dot blot assay (Fig. 3). The difference in the average mycoplasmal load between CFS and non-CFS PCR positive samples was not found to have any correlation with the manifestation or severity of clinical symptomatologies associated with CFS. Treatment with different antimicrobial agents including doxycycline, minocyclin and ciprofloxacin was administered by different clinicians. The clinical efficacy of this treatment should await follow-up study of the bacterial load.
Finally the results of this PCR for M. fermentans in controls and CFS patients should be interpreted carefully until culture isolation demonstrates at least some viable organisms in CFS patients who have higher than 5000 organisms/ml in their blood. Such a demonstration requires an improvement in culture methodologies for better isolation of intracellular organisms such as M. fermentans. Their intimate interaction with the host cells and/or internalization by macrophages and monocytes may prevent their growth. Detection of IgG and IgM antibodies against the M. fermentans lysate and mainly against specific peptides shown in three CFS patients (Table 2) is supportive of preliminary evidence for the specificity of signal detected by this PCR assay, which should be eventually confirmed further by culture isolation.