© 2004 American Thoracic Society
Type 2 Cytokines in Respiratory Syncytial Virus BronchiolitisTo the Editor:We read with great interest the study by Legg and coworkers (1) about immune responses in infants with respiratory syncytial virus (RSV) infection (1). We would like to raise a few points about the data. On the basis of the study's conclusions, infants with bronchiolitis would be expected to have a Th2 bias. However, it was surprising that interleukin (IL)-4 was undetectable by ELISA in 11% of infants with upper respiratory tract infection, but in up to 42% of those with bronchiolitis. Had the proportions (samples with undetectable IL-4) been similar, there may not have been a significant difference in IL-4 between the groups, because in the analysis, the lower limit of assay detection (0.32 pg/ml) was used when IL-4 was undetectable. This may reflect the sensitivity of the assay, as IL-4 is known to be a low copy number cytokine that is active at very low concentrations (2). The cytokine polymerase chain reaction (PCR) data are based on end-point rather than real time PCR. Real time PCR provides a more accurate and reliable quantification over a wide dynamic range (3). There is no reference to PCR reaction cycle number or whether the reaction was in the logarithmic phase, which is essential if end-point PCR is to be used in this way. In our experience, gel densitometric analysis of PCR products provides a dynamic range of approximately 1.5 to maximally 2 logs. Legg and colleagues (1) report the IL-18/glyceraldehyde-3-phosphate dehydrogenase (GAPDH) ratio to change over 4 logs. This would only be possible using this technique if GAPDH was also changing in the opposite direction. Was the housekeeping gene (internal reference) validated for this particular experiment? References like GAPDH that are used to control for error between samples (error of sample-to-sample variation, RNA extraction, reverse transcription, etc.) can be highly variable and prone to directional shifts induced by the underlying study parameter. This was shown in studies of human asthma where GAPDH was not only unsuitable as a reference, but resulted in falsely significant results between patient groups (4). There is now increasing data to suggest that customary housekeeping genes like GAPDH can be highly variable and may cause misinterpretation of study findings (5, 6). Finally it is unclear from the methods how and whether RNA template quality was assessed. This is important because it may account for small but significant group differences. All the above-mentioned confounders may have contributed somewhat to intergroup differences and are especially pertinent when intergroup differences are small but significant.
UCL and Royal Free Medical School London, United Kingdom FOOTNOTES Conflict of Interest Statement: K.D., J.F.H., L.K., and A.Z. have no declared conflict of interest. REFERENCES
From the Authors: We thank Dheda and colleagues for their valuable comments about our article (1). In our study, we showed that respiratory syncytial virus (RSV) bronchiolitis is associated with a profound imbalance in type 1/type 2 cytokines with deficient type 1 and excess type 2 responses. Nasal lavage interleukin (IL)-4 levels detected by ELISA were indeed extremely low. However, in keeping with the study's conclusions, IL-4 was chiefly undetectable in the upper respiratory tract infection (URTI) group (13 of 38 specimens had levels below the detection limit of the assay), whereas the vast majority of bronchiolitis specimens had detectable levels (2 of 18 below the detection limit of the assay). Using the lower limit of detection of the assay (0.32 pg/ml) when IL-4 was undetectable, our analyses would have been more likely to underestimate any difference between the two groups. We thank Dheda and colleagues for giving us the opportunity to make additional comments about the cytokine polymerase chain reaction (PCR) data. To optimize the accuracy of each cytokine PCR, cycle numbers were predetermined to ensure that the reaction was in the logarithmic phase at completion. As highlighted, the IL-18/glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA ratio obtained from stimulated peripheral blood mononuclear cells does indeed vary by more than 4 logs. However, further analysis of our data has revealed that GAPDH levels do not differ significantly between the URTI and viral bronchiolitis groups and do not contribute to the significant difference observed. Furthermore, the gel densitometric analysis used in our study has a maximum linear dynamic range of 5 orders of magnitude (2).
Although traditional housekeeping genes can be highly variable under certain conditions (3), we believe that our data are extremely robust despite the necessity for a housekeeping control for IL-18 and IL-12 mRNA levels. First, a highly significant increase in the IL-4/IFN- Although we did not formally assess the quality of RNA template, all specimens were processed in an identical manner. One million live peripheral blood mononuclear cells (as assessed by trypan blue exclusion) were stimulated with mitogen for 24 hours. Total RNA was then extracted by a standardized technique. A detectable PCR signal was produced by all specimens. We hope that we have addressed all of the issues raised adequately.
University of Southampton Southampton, United Kingdom Imperial College of Science Technology & Medicine London, United Kingdom FOOTNOTES Conflict of Interest Statement: J.P.L. has no declared conflict of interest; I.R.H. has no declared conflict of interest; J.A.W. has no declared conflict of interest; S.L.J. has no declared conflict of interest; J.O.W. has been Chairman of the Scientific Advisory Board for a peanut allergy study employing anti-IgE for Novartis/Genetech and has been a speaker at scientific meetings and courses organized and financed by a variety of pharmaceutical companies including AstraZeneca, Merck, GlaxoSmithKline, and UCB Parma and does not have any financial relationship with a commercial entity that has any interest in the subject of this letter. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ratio was observed in the respiratory tract and systemically. This alone provides strong evidence for an excess type 2 and/or deficient type 1 immune response in infants with bronchiolitis. Second, GAPDH levels did not differ significantly between the two groups and did not contribute to the significant difference in IL-18 mRNA levels observed.