MATERIALS AND METHODS

Clinical Specimens
Endotracheal aspirates from 57 intubated Malaysian premature neonates were collected soon after birth but prior to surfactant therapy, within the first 12 h of admission to the University Malaya Medical Centre (UMMC) Neonate Unit. The samples were collected with informed consent from the neonats’ parents. Endotracheal aspirates were collected during suction of a neonate’s endotracheal tubes by passing a sterile catheter into the distal tracheal and using a non-bacteriostatic saline lavage. The specimens were inoculated into 2 mL of an adapted transport media supplied by the Mycofast^® kit manufacturer and immediately transported to the laboratory for culture identification and PCR assays.

Culture Identification Assay
The presence of M. hormonis and U. urealyticum would be determined by the Mycofast^® Screening Evolution 3 diagnostic kit (International Microbio, France), which relied on the ability of M. hormonis and U. urealyticum to hydrolyse arginine and urea, respectively, in A7 agar. In general, this kit can only be used to culture and detect U. urealyticum and M. hominis. Specimens were processed according to the manufacturer’s recommendations and results were read after incubation for 24 to 48 h at 37 °C.

Nested PCR Assay
Specimen collected were cultured on Mycoplasma Agar (Oxoid, England) with appropriate conditions and incubated at 37 °C for 7 days. The culture was isolated by rinsing with 100 μL of sterile distilled water. The supernatant of the samples were collected and kept at - 70 °C until PCR screening. Five microliter of the sample were used directly as the template in the PCR assay. The presence of various Mycoplasma species and U. urealyticum would be detected using the commercially available PCR Mycoplasma Detection Set (TaKaRa Bio Inc, Japan). The nested PCR reactions contained a 0.1 mM concentration of dNTP mixture, 5 ρmol of each respective MCGp F1 (5’-ACACCATGGGAGCTGGTAAT-3’) and MCGp R1 (5’-CTTCWTCGACTTYCAGACCC AAGGCAT-3’) primers (for the first PCR reaction) and MCGp F2 (5’-GTTCTTTGAAAACTGAAT-3’) and MCGp R2 (5’-GCATCCACCAWAWACTCT-3’) primers (for the second PCR reaction) and 2.5 U of TaKaRa Taq^TM DNA polymerase (TaKaRa Bio Inc, Japan). All reactions in a final reaction volume of 25 μL were performed in a PCR Mastercycler^® gradient (Eppendorf, Germany) under the following conditions: 1 cycle of 30 sec at 94 °C, followed by 35 three step cycles of 94 °C for 30 sec, 55 °C for 60 sec and 72 °C for 60 sec, followed by 10 min at 72 °C. The PCR products were separated on ethidium bromide stained 1.5 % (w/v) agarose gel (Merck, USA). In a separate reaction, each sample of DNA was tested by adding supplied control DNA and running a PCR to determine if amplification inhibitors were present.

RESULTS AND DISCUSSION

No bacterial growth was observed in the Mycofast^® diagnostic kit assay. Meanwhile, the PCR Mycoplasma Detection kit detected neither U. urealyticum nor M. hominis in 57 tested samples, although other Mycoplasma species could be present with an approximately 500 bp band shown in one of the 57 samples (Fig. 1). The second PCR of the nested PCR analysis confirmed that the particular sample might have contained mixed cultures of M. neurolyticum and M. pulmonis (Fig. 2).

Mycoplasma and Ureaplasma, especially M. hominis and U. urealyticum are two common sexually transmitted genital microflora found in sexually active adults (Heggie et al., 2001). Most colonized women are asymptomatic; however, vaginal colonization associated with preterm delivery has been linked to high mortality rates in newborn neonates (Kafetzis et al., 2004; Kataoka et al., 2006). The microbes may be parentally transmitted to the newborns from the mother at the time of delivery or by direct invasion of the fetus in utero (Waites et al., 1993; Lyon, 2000; Pandey et al., 2007).

In recent years, several studies attempted to relate the possibility of an association between M. hominis and U. urealyticum in respiratory colonization with the development of CLD in preterm neonates (Iles et al., 1996; Agarwal et al., 2000; Kafetzis et al., 2004; Kotecha et al., 2004). On the other hand, there have been studies that failed to detect any association (Heggie et al., 2001; Ollikainen et al., 2001; Pandey et al., 2007). Thus, to date, the role of M. hominis and U. urealyticum in the pathogenesis of CLD remains controversial. The great variations in sample selection, sampling methods and identification assays applied might explain the different results observed in those studies.

Endotracheal aspirates are valuable markers of lower respiratory tract infections (Cassell et al., 1998); therefore, it could be beneficial to perform quantitative cultures for detection of Mycoplasma species and U. urealyticum in lower respiratory secretions soon after birth. In the past, a culture would be the only assay for detection of Mycoplasma and Ureaplasma. As a consequence, M. hominis and U. urealyticum were among the least frequently diagnosed respiratory pathogens in clinical specimens.

Fig. 1:	First PCR result of nested PCR assay for identification of Mycoplasma species and Ureaplasma urealyticum. Lane 1: 100-bp ladder (Fermentas, Maryland); Lanes 2 and 4: tested samples; the ~500 bp band in Lane 2 indicates the presence of Mycoplasma species; Lanes 3 and 5: tested samples with added internal control DNA (~ 810 bp); Lane 6: DNA blank (sterile distilled water)

Fig. 2:	Second PCR result of nested PCR assay. Lane 1, 100-bp ladder (Fermentas, Maryland); Lane 2, DNA blank; Lane 3, Result after PCR using amplicon obtained from first PCR. Band with 196 bp indicated the presence of M. neurolyticum and 189 bp for the M. pulmonis

In addition, culture identification is slow; it can take up to 2-5 days to culture Ureaplasma and M. hominis and possibly up to 8 weeks to culture M. genitalium (Waites et al., 2005). Historically, culture identification is considered definitive; colony identification is difficult, subjective and lacks sensitivity, since it is based on observation under a light microscope. It is also vulnerable to cross-contamination with other microorganisms that might generate false-positive results.

Now, with advancements in molecular techniques, there have been evaluations of different PCR systems, commercial as well as in-house PCR assays, that showed high sensitivity, specificity, rapidity and the prospect of direct identification of many Mycoplasma species as well as Ureaplasma species in clinical specimens (Abele-Horn et al., 1996; Luki et al., 1998; Stellrecht et al., 2004; Dhawan et al., 2006; Hashimoto et al., 2006). A PCR assay would have better diagnostic potential compared to cultures in the detection of Mycoplasma and Ureaplasma in neonatal respiratory samples. The enhanced sensitivity and specificity of the PCR also suggest that PCR methods could be suitable as a routine diagnostic tool in identification of Mycoplasma and Ureaplasma, especially in diagnostic laboratories that do not currently test for Mycoplasma and Ureaplasma.

In the present study, Mycofast Screening kit was unable to detect other Mycoplasma and Ureaplasma species besides M. hominis and Ureaplasma urealyticum and this could explain the failure of detection of other Mycoplasma species in one of our specimens by using this kit. However, we used a second approach, PCR to screen all the samples again. Commercially available PCR Mycoplasma Detection Set used in this study relied on the principle of PCR for rapid identification of various Mycoplasma species and U. urealyticum (Uemori et al., 1992). From the results obtained, neither M. hominis nor U. urealyticum was detected in any of the 57 tested samples. Only one of 57 specimens was found to carry mix infection of M. neurolyticum and M. pulmonis. In general, false-negative results might have occurred if the specimens contained PCR amplification inhibitors; however, our samples with additional control DNA shown absence of inhibitors with all samples generating an approximately 810 bp amplicon (Fig. 1) in the internal control test.

Generally, inability to detect Ureaplasma and Mycoplasma by either serological or molecular method might also be attributed to the length of time the tracheal aspirate specimens had been stored; a period of several months to a year before laboratory analysis began. Tracheal aspirates were collected in normal saline only, without being inoculated in any storage media. Some degree of deterioration may have occurred to a point where no bacteria could be detected at all. Therefore, analysis should be carried out as soon as possible once the samples are collected.

In conclusion, as only a single sample was detected to have a mix culture of Mycoplasma species by means of the nested PCR approach, this did not give a significant association between U. urealyticum and Mycoplasma species infection with CLD. Therefore, we conclude CLD in Malaysian preterm neonates is not due to colonization of U. urealyticum or Mycoplasma species in their respiratory tract.