The results of this study suggest a strong degree of correlation between the abilities of two different laboratories to detect the presence of C. pneumoniae
in the CSF from MS patients but not from OND controls. These results also show that these two laboratories were able to detect C. pneumoniae
at levels that were significantly higher in MS patients than in OND controls. The two laboratories used different techniques for DNA extraction and for PCR testing. The VUMC laboratory modified the lysis buffer and used phenol-chloroform for DNA extraction. A nested-PCR assay with two different sets of primers, both of which are specific for the MOMP gene of C. pneumoniae
, were used. While the Tong and Sillis primers have been used before for the detection of C. pneumoniae
in clinical specimens, the running conditions for the present study were modified to detect C. pneumoniae
in CSF at concentrations of 0.05 IFU/μl (17
). Similar strategies were also employed prior to the use of the VU-MOMP primers.
The USF laboratory performed DNA extraction with the Qiagen DNA mini kit with bacterial DNA extraction protocols. The primer set that the USF laboratory used was developed and previously reported to be sensitive for the detection of chlamydial DNA at 1 IFU per PCR mixture. Since the IFU number shows only the number of infective EB, which may be affected by the protocols of EB isolation and culture conditions, the absolute PCR sensitivities for protocols based on IFU numbers prepared in different institutes are difficult to compare. The results of this comparative study make it unlikely that the PCR product detected in CSF from MS patients was an artifact of DNA amplification.
Since our initial report on the association between the presence of C. pneumoniae
in the CSF of MS patients and the development of MS, there have been a number of studies that have attempted to reproduce these observations. Most of these studies have met with limited success. Layh-Schmitt et al. (8
) was able to show the presence of the organism in 22% of MS patients, and this number increased when phenol-chloroform extraction techniques were used (17
). More recently, Contini and Fainardi (41st ICAAC) have reported the presence of C. pneumoniae
in the CSF of 58% of MS patients, and another report (15
) noted the presence in 10% of MS patients. It should be noted that in all these studies, the presence of the C. pneumoniae
genome was statistically higher in MS patients than in control patients. In a report by Gieffers et al., the presence of C. pneumoniae
in the CSF was demonstrated in a high number of OND patients by PCR (4
). This OND group included a number of individuals with inflammatory disease, including optic neuritis. Therefore, these patients might represent patients with early symptoms of MS. Investigators from the University Hospital of Umeå could not document the presence of C. pneumoniae
by PCR in any CSF samples previously collected from 48 MS patients as well as in 51 OND patients (1
). Other groups also have failed to identify C. pneumoniae
by PCR in CSF samples and the tissues of patients with MS (13
; Marcos et al., presented at the International Chlamydia
Meeting, Helsinki, Finland, 2000).
In a recent collaborative study involving four different laboratories, the presence of C. pneumoniae
was examined in 52 blinded CSF samples (6
). VUMC was able to detect C. pneumoniae
in 73% (22 of 30) of MS patients compared to 22% (5 of 22) of controls. Three other laboratories (C. Gaydos, Johns Hopkins University, Baltimore, Md.; J. Boman, Umeå, Sweden; and L. Tondilla, Centers for Disease Control and Prevention, Atlanta, Ga.) failed to show the presence of C. pneumoniae
in any of the CSF samples. This study also suggests that the presence of the organism is not due to any regional differences in the patient population, since all samples were obtained from Carolinas Medical Center, Charlotte, N.C. The procedures used by these four laboratories were different, and since the samples were blinded with regard to the underlying diagnosis, technical differences could very likely account for the differences in the outcomes of the PCR studies. The ability of the laboratory at USF to validate results obtained from VUMC suggests that technical problems in the PCR procedure (which may include sample collection and storage as well as extraction of DNA and PCR testing) might account for the discrepancies among results from different laboratories.
The preliminary study conducted at USF and published earlier had shown that optimization of the PCR conditions is necessary for obtaining a signal for C. pneumoniae
in the presence of low copy numbers of the organism (5
). These issues were extensively reviewed by us, and the importance of DNA extraction techniques, primer selection, and optimization or running conditions have been previously discussed (Zuzak et al., presented at the International Chlamydia
Meeting, Helsinki, Finland, 2000). It is clear, for example, that primers for MOMP are more sensitive than the 16S RNA gene for the detection of C. pneumoniae
. This may explain the inability of the C. Gaydos laboratory to detect C. pneumoniae
in the CSF of MS patients (6
In our previous studies, we had amplified the entire 1.1-kb MOMP gene. In the present study, we used the Tong and Sillis primers, since they have been used extensively to document the presence of C. pneumoniae
, in addition to new primers that produced less background signal. The laboratory of J. Boman has also used the Tong and Sillis MOMP primers but has relied on the Qiagen blood mini kit for DNA extraction (6
). We were unable to detect a signal for C. pneumoniae
in the CSF of MS patients with the Tong and Sillis primers following DNA extraction with the Qiagen blood mini kit. We therefore believe that DNA extraction may prove more efficient when performed with phenol-chloroform or with the QIAmp bacterial DNA extraction kit.
Although we have not quantitated the number of copies of the C. pneumoniae
gene that is present in the CSF of MS patients, it is very likely that this copy number is low. Others have shown that performance of PCR in triplicate or more, as well as performance of probit analysis, may be necessary to exclude the problems of dilution and statistical variability in the presence of low copy numbers (14
Importantly, our study also suggests that CNS infection with C. pneumoniae
occurs early in the course of the disease, since the organism was identified in 50% of MS patients who presented heralding signs of MS but who had not yet shown dissemination of the disease. In situations wherein patients present with clinically isolated syndromes that are highly characteristic of MS (e.g., optic neuritis, internuclear ophthalmoplegia, or partial transverse myelopathy), the definitiveness of the diagnosis is uncertain, since the essential criteria of the dissemination of lesions over time has not yet been satisfied. Long-term studies of patients with isolated clinical syndromes suggest that the initial abnormalities on magnetic resonance images (MRI) and the presence of oligoclonal bands are likely to predict conversion to clinically definite MS (2
). These prospective studies suggest that at 15 years, 88% of patients presenting with clinically isolated syndromes suggestive of MS progress to clinically definite MS (2
). Although the number of patients in our study was small, three patients (patients 1, 2, and 4) who were positive by PCR for the presence of the C. pneumoniae
gene and one patient (patient 3) who was negative by PCR for C. pneumoniae
have subsequently progressed into clinically definite MS.
In conclusion, our study further substantiates our initial observation regarding the increased presence of C. pneumoniae genes in the CSF of MS patients relative to that of controls. Also, the presence of C. pneumoniae in patients with signs of MS indicates that infection by this microorganism in MS occurs early in the course of the disease. The significance of this observation in relation to the role of C. pneumoniae in the pathogenesis of MS remains uncertain. We believe that a clinical trial that monitors the course of MS during and after eradication of the organism is most likely to explain conclusively the causal nature of this association.