Real-time PCR for quantification of Helicobacter felis in mouse stomach.

To the Editor: Helicobacter pylori is the most common cause of chronic bacterial infections, affecting at least 50% of the world's population and leading to gastritis, peptic ulcer disease and gastric adenocarcinoma. Development of a vaccine to control this infection has become a priority, and preclinical testing of potential products includes inoculation of mice with H felis and determination of the number of bacteria in the stomach. (1) Since the traditional quantitative culture technique is laborious, (2) we--as well as others (3,4)--have developed a real-time polymerase chain reaction (Q-PCR) for this task. Here, we describe our assay.

We used the iCycler iQ Multicolor Real-Time system (Bio-Rad Laboratories, Hercules, CA) for amplification and detection of the fluorescent signal, with the following run conditions: one cycle of 95[degrees]C for 13:30 minutes, 40 cycles of 15 seconds at 94[degrees]C, 30 seconds at 55[degrees]C and 30 seconds at 72[degrees]C. Specimens were processed in duplicate, with each reaction being 25 [micro]L and including 12.5 [micro]L iQ-Supermix (Bio-Rad), 1 [micro]L each of sense and antisense primer, 0.5 [micro]L of probe (200 nM for all three) and 1 [micro]L of test DNA. The set of sense primer (5'-TAT CGC TAA AGG ATT GGT CTA TG-3'), antisense primer (5'-CGG ATA CCC GTC ATA GCC-3'), and probe (5'-/56-FAM/AGC CTT TAC CTC ACC AAC AAG CTG/3BHQ-1/-3') target a 77 bp stretch of the 16S ribosomal RNA gene of H felis (GenBank #: U51872). Standards for the assay were prepared by cloning the H felis gene into the pGEM T vector (Promega Corp., Madison, WI), replicated in Escherichia coli and serial 10-fold dilutions of purified plasmid DNA subjected to Q-PCR. Plotting of [log.sub.10] concentration of each specimen and its corresponding mean Ct reading showed a linear correlation ([r.sup.2] = 0.98) for concentrations between [10.sup.10] to 10 copies/[micro]L (Fig.). We then tested the Q-PCR in Helicobacter DNA. For this, we cultured H felis (ATCC 49179 strain) and H pylori (SS1 strain, kindly provided by Dr. Yoshio Yamaoka, Baylor College of Medicine, Houston, TX), purified total DNA, prepared serial 10-fold dilutions, and submitted to Q-PCR. The assay was sensitive and accurate to detect H felis at concentrations [10.sup.6] to 10 copies/[micro]L, while for H pylori, the signal was weak, about 1,000-fold lower than for H felis and not detectable for levels [less than or equal to] [10.sup.4] copies/[micro]L, indicative of lower specificity of the primers for H pylori as compared with H felis.

For our purposes, the most important aspect is the performance of Q-PCR in mouse stomach since that is its intended use. For this, total DNA was purified from the stomachs of 11 mice: 1 not inoculated and 10 inoculated with H felis. The uninoculated stomach was negative by Q-PCR (85 copies/[micro]g DNA), while all the inoculated specimens were positive (range: 76,631-493,311 copies/[micro]g). To determine the reproducibility of the assay, stomach specimens with concentrations in the low (17,599 copies/[micro]g), middle (59,979 copies/[micro]g) and high (116,950 copies/[micro]g) spectrum of Q-PCR readings were retested three more times in duplicate. The intra- and inter-run variability at all three levels tested (as measured by the coefficient of variation = 100 X standard deviation / mean) was 1.7 to 3.6% and 0.8 to 5.7%, respectively, for the [log.sub.10] concentration, which corresponds to 10 to 40% variability when expressed as the transformed (antilog) number of copies/[micro]L. By comparison, light microscopy identified H felis in only 5 of the 10 inoculated stomachs, suggesting that Q-PCR is more sensitive than light microscopy to detect H felis infection. Indeed, light microscopy has been described as insensitive to detect Helicobacter infection levels below [10.sup.5] bacteria/g tissue. (5)

In conclusion, we describe a Q-PCR assay that is rapid, easy to perform, sensitive, specific for H felis and reproducible. We believe this method is of potential utility for the evaluation of H pylori vaccine candidates and other experimental models of H felis infection.

Acknowledgments

We thank W. Douglas Scheer, PhD from LSUHSC, New Orleans, LA, for his advice on PCR techniques.

Rodolfo E. Begue, MD

Jennifer Manning, BSN

Hernan Correa, MD

Division of Infectious Diseases and Nephrology

Departments of Pediatrics and Pathology

Louisiana State University Health

Sciences Center

New Orleans, LA

References

1. Del Giudice G, Covacci A, Telford JL, et al. The design of vaccines against Helicobacter pylori and their development. Annu Rev Immunol 2001;19:523-553.

2. Atherton JC, Tham KT, Peek RM, et al. Density of Helicobacter pylori infection in vivo as assessed by quantitative culture and histology. J Infect Dis 1996;174:552-556.

3. Jiang W. Baker HJ, Smith BF. Mucosal immunization with Helicobacter, CpG DNA, and cholera toxin is protective. Infect Immun 2003;71:40-46.

4. Stoicov C, Whary M, Rogers AB, et al. Coinfection modulates inflammatory responses and clinical outcome of Helicobacter felis and Toxoplasma gondii infections. J Immunol 2004;173:3329-3336.

5. Sutton P, Wilson J, Lee A. Further development of the Helicobacter pylori mouse vaccination model. Vaccine 2000;18:2677-2685.

Concentration  Ct-1   Ct-2   Ct-mean

10^10           9.30   9.00   9.15
10^9           12.40  11.30  11.85
10^8           12.60  11.70  12.15
10^7           15.80  15.40  15.60
10^6           22.00  21.50  21.75
10^5           22.90  22.30  22.60
10^4           27.30  26.20  26.75
10^3           29.70  29.10  29.40
10^2           33.10  32.40  32.75
10^1           35.30  33.70  34.50
10^0           NA     35.70  35.70
Blank          NA     36.80  36.80

Fig. Development of the standard curve by Q-PCR of 10-fold dilutions of
pGEM/H felis plasmid. Linear amplification ([r.sup.2] = 0.9847) for the
[log.sub.10] concentration from [10.sup.10] to [10.sup.0] copies per
reaction. Blank specimen had water (no DNA) added; NA = no
amplification.