During 2006–2009, the 267 samples were collected from 3 groups of patients (). Group 1 comprised patients who had traveled from Asia to Germany; location of their permanent residency was unknown. Groups 2 and 3 and the control group comprised only persons from northern Germany. From group 1, a total of 75 sputum and nasal swab specimens were collected from patients who had unconnected cases of RTI and who had recently traveled by air (11
). From group 2, a total of 31 bronchoalveolar lavage (BAL) samples were collected from patients with chronic obstructive pulmonary disease (defined by a forced expiratory volume in 1 second/forced vital capacity <70% and forced expiratory volume in 1 second <80% of the predicted value) who had signs of RTI. From group 3, a total of 161 BAL and tracheal secretion samples were collected from patients with severe RTI and immunosuppression as a result of solid organ or bone marrow transplantation. From the control group, throat swabs were collected from 52 healthy persons and BAL samples were collected from 10 healthy volunteers who had no signs of RTI and no known underlying disease.
Detection of XMRV in respiratory tract secretions from 329 persons*
All samples were analyzed by culture for pathogenic bacteria and fungi and by PCR for rhinoviruses, adenoviruses, enteroviruses, influenza viruses A and B, parainfluenza viruses 1–3, respiratory syncytial virus, cytomegalovirus, Epstein-Barr virus, and human metapneumovirus. All samples were tested in duplicates obtained by individual RNA extractions. XMRV RNA was reverse transcribed from total RNA, after which nested PCR or real-time PCR were conducted as recently described (1
). No serum samples were available from these patients to confirm the results by serologic testing.
For group 1, XMRV-specific sequences were detected with relatively low frequency (2.3%). For group 2, XMRV-specific sequences were amplified in 1 BAL sample, which was also positive for Staphylococcus aureus by routine culture methods. For group 3, XMRV-specific sequences were detected at a frequency of 9.9%, which was significantly higher than that for the healthy control group (3.2%) at the 90% confidence level but not at the 95% level (p = 0.078, 1 sample t-test). Of 16 group 3 samples, 10 showed no signs of co-infection. The remaining 6 samples showed co-infection with rhinovirus or adenovirus (1 sample each); S. aureus (3 samples); or mixed infection with pathogenic fungi, Candida albicans and Asperigillus fumigatus (1 sample).
All samples that were positive for XMRV by gag-nested PCR, together with a set of those that were negative for XMRV, were retested by real-time PCR. Results showed low XMRV RNA concentrations, 103 –104/mL of specimen.
To confirm the validity of XMRV detection, a subset of 6 specimens (3 XMRV positive and 3 XMRV negative) were tested by using an alternative PCR assay for viral RNA (3
) and a C-Type RT Activity Kit (Cavidi, Uppsala, Sweden) for type C reverse-transcription activity. XMRV sequences from alternative targets in the gag and env regions were confirmed in 2 of the 3 XMRV-positive samples but in none of the controls. One XMRV-positive BAL specimen showed an 8-fold increase above background of specific type C retroviral reverse-transcriptase activity, suggesting presence of active type C retrovirus within this sample. This assay is substantially less sensitive than reverse transcription–PCR.
All XMRV gag sequences (390-bp fragment) were 98%–99% identical to previously published XMRV sequences from persons with prostate cancer (1
). Phylogenetic analysis showed close clustering ().
Figure Xenotropic murine leukemia virus–related gammaretrovirus (XMRV) gag sequences derived from respiratory tract secretions. Phylogenetic tree comparing the 390-nt gag fragment of all respiratory samples of this study with recently published XMRV (more ...)