Abstract
We validated the use of stored samples for Chlamydia trachomatis research. C. trachomatis DNA was detected by real-time PCR in clinical (urine and self-taken vaginal swabs) and spiked samples using six different media, five different time points (up to 2 years), and four different temperature conditions. C. trachomatis was detected in all 423 samples, and no clinically relevant degradation impact was detected.
TEXT
Chlamydia trachomatis is the most prevalent bacterial sexually transmitted microorganism worldwide. Many researchers conveniently use stored samples for their C. trachomatis research. In previous decades, the viability of C. trachomatis has been extensively explored using culture, and freezing samples appeared to be the most essential step in keeping C. trachomatis culturable (1, 2). The package insert of the COBAS TaqMan CT test (3), for instance, states that urine can be stored refrigerated or frozen for at maximum 7 and 30 days, respectively, before being processed. Swabs in transport medium can be stored at room temperature and frozen for, at maximum, 14 and 30 days, respectively, according to the package insert (3). The effect of different storage conditions on the load of C. trachomatis using nucleic acid amplification tests (NAAT), however, has never yet been thoroughly assessed in a clinical trial. We hypothesized that storage would not lead to false-negative NAAT results. Therefore, we assessed the impact of four different temperature conditions, six different types of medium, and five increasing lengths of duration of storage, up to 2 years, on C. trachomatis DNA detection.
For this purpose, phosphate-buffered saline (PBS), 2-sucrose-phosphate (2-SP) medium, COBAS Amplicor medium (Roche Diagnostics, Mannheim, Germany), and urine samples were spiked with the same amount of C. trachomatis serovar D elementary bodies (MyBioSource, San Diego, CA) and were stored at room temperature (RT), 4°C, −20°C, and −80°C, in triplicate. Samples were thawed only once, on the day of C. trachomatis DNA testing. Furthermore, clinical C. trachomatis-positive urine samples, as well as C. trachomatis-positive swabs in COBAS Amplicor medium, were collected, pooled, and stored in triplicate at the same four temperatures.
Samples were tested in triplicate on day 0 and after 1, 7, 14, and 30 days and 2 years of storage (136 clinical and 287 spiked samples) for the presence of C. trachomatis DNA. DNA was isolated using the Qiagen DNA minikit (Qiagen GmbH, Hilden, Germany). For the real-time PCR targeting the cryptic plasmid as described by Jalal et al. (4), the total PCR volume was adjusted to 50 μl. Also, only the inner primers were used, to avoid a nested PCR setup. For PCR amplification, an ABI 7900 HT real-time PCR machine (Applied Biosystems, Carlsbad, CA) was employed. Approximately 3,000 plasmids were available per PCR (e.g., per 20-μl sample used in the PCR).
Generalized linear models were used, controlling for repeated measurements. Models were run separately for the six evaluated modalities of samples. We tested whether the number of PCR cycles needed to detect C. trachomatis DNA changed as storage time increased. An increase in the number of cycles needed corresponds to a decrease in C. trachomatis DNA detected compared to the amount in the previous sample. An increase of 3.3 PCR cycles corresponds to an approximately 1-log decrease in C. trachomatis DNA load. Furthermore, the influence over time of storage temperature, with four categories, was examined for the different media. Due to the large time interval, generalized linear models were only used for analyzing data obtained within the first month. Analyses were conducted using SPSS19 (IBM Corporation, Somers, NY); a P value of <0.05 was considered statistically significant.
C. trachomatis could be detected in all clinical samples and spiked media at all time points and irrespective of the storage temperature (Table 1). For spiked PBS and 2-SP and pooled C. trachomatis-positive swabs in COBAS Amplicor medium, the cycle threshold was independent of storage duration and temperature within the first month. For C. trachomatis DNA detection in spiked COBAS Amplicor medium, the cycle threshold increased within the first month at −20°C and −80°C (both P < 0.01), while time trends showed a nonsignificant (P = 0.09) decrease at room temperature and stability at 4°C (P = 0.95). Finally, for spiked urine and for pooled clinical urine samples, the cycle threshold decreased within the first month (P < 0.01), including all but one (4°C, P = 0.09) of the studied temperatures, reflecting an increase in C. trachomatis DNA load (data not shown). Regarding the results obtained after the 2-year storage interval, several findings are noteworthy (Table 1). The cycle thresholds in the spiked PBS and 2-SP experiments were stable over time. For spiked COBAS Amplicor medium, the cycle threshold, which had increased during the first month, was found to have decreased in the samples frozen for 2 years. In both spiked and clinical urine samples, the cycle threshold had increased after 2 years in the frozen samples, after the initial decrease.
Table 1.
Cycle threshold values of Chlamydia trachomatis DNA in various media at different time pointsa
| Specimen | Temp (°C) |
CT value [average (SD)] at indicated time point |
|||||
|---|---|---|---|---|---|---|---|
| Day(s) |
2 yr | ||||||
| 0 | 1 | 7 | 14 | 30 | |||
| Spiked PBS | RT | 27.90 (1.59) | 28.50 (0.37) | 29.35 (1.90) | 27.54 (0.40) | 27.67 (0.11) | 27.15 (0.59) |
| 4 | 29.02 (0.18) | 28.76 (0.22) | 29.03 (0.04) | 28.81 (0.41) | 27.88 (0.12) | 28.48 (0.36) | |
| −20 | 29.58 (0.58) | 28.87 (0.63) | 31.29 (0.28) | 29.58 (0.69) | 29.28 (0.33) | 29.71 (0.16) | |
| −80 | 29.10 (0.81) | 28.22 (0.69) | 28.00 (2.14) | 27.88 (0.97) | 28.68 (1.31) | 29.30 (0.25)c | |
| Spiked 2-SP | RT | 24.93 (0.67) | 24.03 (0.23) | 24.44 (0.73) | 25.04 (1.02) | 24.41 (0.23) | 23.03 (0.44) |
| 4 | 24.59 (1.00) | 23.42 (0.64) | 24.21 (0.97) | 24.10 (0.27) | 23.80 (0.90) | 23.34 (0.73) | |
| −20 | 24.82 (0.83) | 24.59 (0.84) | 25.19 (1.58) | 25.39 (0.75) | 23.08 (1.32) | 23.62 (0.67) | |
| −80 | 25.40 (0.41) | 24.78 (1.32) | 24.23 (1.16) | 25.22 (0.49) | 24.45 (0.47) | 23.92 (0.46) | |
| Spiked COBAS medium | RT | 26.00 (0.34) | 25.00 (0.15) | 24.39 (0.95) | 24.42 (0.25) | 24.23 (0.43) | 24.45 (0.44) |
| 4 | 25.30 (0.92) | 23.86 (1.27) | 28.85 (1.48) | 28.96 (0.02) | 24.66 (0.96) | 25.03 (0.52) | |
| −20 | 25.45 (1.00) | 24.70 (1.51) | 28.77 (0.96) | 29.35 (0.72) | 28.54 (0.89) | 23.99 (1.56) | |
| −80 | 25.62 (0.56) | 24.24 (0.83) | 28.71 (0.20) | 29.33 (0.46) | 28.97 (0.86) | 24.37 (1.24) | |
| Spiked urine | RT | 27.85 (1.34) | 25.96 (2.53) | 24.63 (0.75) | 23.34 (0.64) | 23.56 (1.05) | 22.58 (0.17) |
| 4 | 28.78 (2.21) | 26.84 (1.88) | 23.40 (2.17) | 24.19 (0.52) | 24.92 (0.35) | 24.93 (1.78) | |
| −20 | 29.87 (0.65) | 26.61 (0.66) | 24.31 (0.62) | 23.75 (1.37) | 23.61 (1.20) | 32.56 (1.03) | |
| −80 | 29.40 (1.21) | 25.84 (0.95) | 24.22 (0.61) | 24.60 (0.52) | 24.69 (0.87) | 31.84 (2.45) | |
| Clinical C. trachomatis-positive urine | RT | 33.63 (0.75) | 33.21 (1.07) | 30.70 (0.43) | 29.94 (0.19) | 29.97 (0.36) | 30.37 (0.50) |
| 4 | 32.83 (0.87) | 33.73 (0.37) | 34.17 (0.65) | 31.88 (0.83) | 30.71 (0.49) | 29.97 (0.21) | |
| −20 | 33.46 (0.66) | 33.48 (1.08) | 33.59 (0.94) | 32.90 (0.44) | 32.60 (0.24) | 37.16 (0.61) | |
| −80 | 33.81 (0.74) | 33.78 (0.18) | 33.35 (0.84) | 33.00 (0.29) | 32.43 (0.26) | 36.53 (0.23) | |
| Clinical C. trachomatis-positive swabs in COBAS medium | RT | 27.06 (0.48) | 26.19 (0.05) | 27.12 (0.11) | 26.79 (0.37) | 27.39 (0.16) | 34.11 (0.57)c |
| 4 | 26.25 (0.41) | 26.12 (0.19) | 26.79 (0.46) | 26.46 (0.38) | 26.47 (0.20) | 29.89 (NA)b | |
| −20 | 27.32 (0.36) | 25.90 (0.30) | 26.25 (0.10) | 26.50 (0.20) | 26.94 (1.20) | 26.81 (NA)b | |
| −80 | 27.72 (0.49) | 25.84 (0.08) | 26.08 (0.51) | 26.87 (0.67) | 26.07 (0.13)c | 26.34 (NA)b | |
CT, cycle threshold; PBS, phosphate-buffered saline; 2-SP, 2-sucrose-phosphate medium; COBAS, COBAS Amplicor medium; RT, room temperature.
Tested singly.
Tested in duplicate.
C. trachomatis DNA could be detected in all clinical samples and spiked media tested, implying that none of the conditions had a clinically relevant degrading impact on the available C. trachomatis DNA. Nevertheless, several remarkable findings ought to be highlighted.
Although the C. trachomatis DNA input in all spiked samples is similar, variation exists in the test results on day 0 (immediately after composing the samples). It is likely that lysis already had started in the 2-SP and COBAS Amplicor samples, which explains the lower cycle thresholds in these samples in comparison with those of the spiked PBS samples. Since the pooled clinical urine and swab samples contained an unknown C. trachomatis load, their numbers of PCR cycles needed to detect C. trachomatis on day 0 or any other time point cannot be compared directly with the numbers for the spiked samples.
We found a significant decrease in the number of cycles needed over time to detect C. trachomatis DNA in the spiked urine samples within the first month. This decrease did not hold for the frozen samples after 2 years of storage. The stability of C. trachomatis DNA in urine samples was previously investigated by Morré et al. (5). Overall, C. trachomatis could be detected in all their samples, although initial freezing appeared to impair detection. This could have been due to the subsequent release of cellular DNase, as hypothesized by Morré et al. In contrast, in our study, during which the stored samples were not prefrozen, we were able to detect higher loads of C. trachomatis DNA over time in stored urine samples. This might reflect a decrease in the amount of PCR-inhibiting substances known to be present in urine samples (6). Long-term freezing, however, seems to result in degradation of C. trachomatis DNA in urine samples, since the phenomenon was observed for spiked as well as clinical urine samples. As can be seen in Table 1, variations in cycle threshold exist between triplicate measurements in the spiked urine samples (especially at RT and 4°C), which are not present in the clinical urine samples. This could reflect a difference in degrading enzymes and/or inhibiting substances between patients' urine samples. In spiked COBAS Amplicor samples, we detected a significant decrease in C. trachomatis load within the first month. This was not the case in the pooled C. trachomatis-positive swabs in COBAS medium nor in any of the other experiments. The decrease in C. trachomatis DNA load occurred only in the frozen samples, not in samples stored at 4°C or RT. This finding is remarkable, since if the decrease in C. trachomatis DNA load had been the result of an enzymatic process, one would expect this to occur in the samples stored at 4°C or RT, which was not the case. Surprisingly, the C. trachomatis DNA load was found to reverse to initial values in these samples after 2 years of frozen storage. The COBAS Amplicor package insert does not recommend storage of swab samples in COBAS medium.
Maass et al. explored the viability of Chlamydophila pneumoniae after storage in different media and temperatures (7). They found a higher survival rate of C. pneumoniae when samples were frozen at −75°C than at 4°C or 22°C. Eley et al. used different lyophilized C. trachomatis strains which were stored at different temperatures to assess the viability after 1 week and 1 month (8). Storage temperature affected viability, but recovery was relatively stable between the two time points. A rise in temperature clearly affected viability, with no recovery at the highest temperature (37°C) in the study by Eley et al. Using PCR, Chlamydia spp. do not need to be viable and our results indicate that PCR results are significantly less affected. Catsburg et al. described that even C. trachomatis DNA preserved on dry vaginal swabs which were stored at −80°C could be detected after 1 year of storage (9).
Our results demonstrate that storage conditions and duration hardly affect C. trachomatis DNA detection by PCR in a negative manner, although frozen urine samples, stored for prolonged periods (more than 2 years), could become C. trachomatis negative. Nevertheless, our study does validate the use of stored samples in C. trachomatis research. Furthermore, it justifies the use of mailed samples in large screening programs in countries with moderate climate (10) and could be of use in home-based or outreach-based diagnostic testing procedures.
Footnotes
Published ahead of print 19 December 2012
REFERENCES
- 1. Mahony JB, Chernesky MA. 1985. Effect of swab type and storage temperature on the isolation of Chlamydia trachomatis from clinical specimens. J. Clin. Microbiol. 22:865–867 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Tjiam KH, van Heijst BY, de Roo JC, de Beer A, van Joost T, Michel MF, Stolz E. 1984. Survival of Chlamydia trachomatis in different transport media and at different temperatures: diagnostic implications. Br. J. Vener. Dis. 60:92–94 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Roche Molecular Diagnostics 2009. COBAS TaqMan CT Test, v2.0, package insert. Roche Molecular Diagnostics, Basel, Switzerland: http://molecular.roche.com/assays/Pages/COBASTaqManCTTestv20.aspx [Google Scholar]
- 4. Jalal H, Stephen H, Curran MD, Burton J, Bradley M, Carne C. 2006. Development and validation of a rotor-gene real-time PCR assay for detection, identification, and quantification of Chlamydia trachomatis in a single reaction. J. Clin. Microbiol. 44:206–213 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Morré SA, van Valkengoed IG, de Jong A, Boeke AJ, van Eijk JT, Meijer CJ, van den Brule AJ. 1999. Mailed, home-obtained urine specimens: a reliable screening approach for detecting asymptomatic Chlamydia trachomatis infections. J. Clin. Microbiol. 37:976–980 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Huggett JF, Novak T, Garson JA, Green C, Morris-Jones SD, Miller RF, Zumla A. 2008. Differential susceptibility of PCR reactions to inhibitors: an important and unrecognised phenomenon. BMC Res. Notes 1:70 doi:10.1186/1756-0500-1-70 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Maass M, Dalhoff K. 1995. Transport and storage conditions for cultural recovery of Chlamydia pneumoniae. J. Clin. Microbiol. 33:1793–1796 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Eley A, Geary I, Bahador A, Hakimi H. 2006. Effect of storage temperature on survival of Chlamydia trachomatis after lyophilization. J. Clin. Microbiol. 44:2577–2578 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Catsburg A, van Dommelen L, Smelov V, de Vries HJ, Savitcheva A, Domeika M, Herrmann B, Ouburg S, Hoebe CJ, Nilsson A, Savelkoul PH, Morré SA. 2007. TaqMan assay for Swedish Chlamydia trachomatis variant. Emerg. Infect. Dis. 13:1432–1434 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Gotz HM, van Bergen JE, Veldhuijzen IK, Hoebe CJ, Broer J, Coenen AJ, de Groot F, Verhooren MJ, van Schaik DT, Richardus JH. 2006. Lessons learned from a population-based chlamydia screening pilot. Int. J. STD AIDS 17:826–830 [DOI] [PubMed] [Google Scholar]