2 January 2002

Dear Editor

Waning of vaccine-mediated immunity is one of the reasons for measles epidemics in immunized populations. While lifelong protection conferred by natural measles has been explained by the persistence of measles virus (MV) or its antigen in the host, this phenomenon has not been described after immunization. In our investigation of MV RNA and antigen in asymptomatic subjects’ lymphoid tissues and lymphocytes, we observed MV material in the pharyngeal tonsils of some children who had no history of, or recent contact with, measles infection.

Fresh lymphoid tissue samples consisting of pharyngeal tonsils extracted for chronic tonsillitis (n = 10; ages 5−11 years) or abdominal lymph nodes excised in gynecological surgery (n = 5; ages 32−59 years) were examined. The history of measles infection or immunization was available and serum measles antibodies were detectable (>1/4 by the complement fixation method) in all patients. RNA extracted from tissues was subjected to nested reverse transcriptase-polymerase chain reaction (RT-PCR) using the outer DNA primers NP1 and NP4 and the inner primers NP2 and NP3.1 The same tissues were also examined by immunohistochemistry, with antimeasles rabbit serum and peroxidase-conjugated secondary antibody. Sections of an autopsy case with MV-giant cell pneumonia were used as a positive control, and sections incubated with irrelevant primary antibody were used as a negative control in all experiments. Staining intensity was evaluated by a blinded examiner as strong (+++), moderate (++), weak (+), or absent. The study was approved by the Ethics Committee of Hacettepe University Faculty of Medicine.

Out of 15 tissue extracts, four were positive for MV RNA (Fig. 1). These were from the tonsillectomy material of children 5 to 7 years of age with chronic tonsillitis. Two of these tissues were also weakly positive by immunohistochemistry. All MV RNA-positive children had serum measles antibodies at a titre of 1/4; they had no history of measles infection, and two of them had a definite history of immunization for measles at the age of 15−18 months. Out of the five lymph nodes, two were positive by immunohistochemistry, but negative by RT-PCR.


Figure 1. Reverse transcriptase-polymerase chain reaction: lanes 1, 2, 4 and 10 are positive for the N protein gene of MV on a 2.5% agarose gel (arrow). The size marker is phage X174 DNA digested with HaeIII.

Download figure to PowerPoint

Wild MV RNA has previously been found in the peripheral blood lymphocytes of healthy donors, or in non-immunized children exposed to MV. Only wild-type viruses are amplified from human tissues, unless the subject has recently been vaccinated.2,3 Because the children in the current study had been immunized more than 2 years ago and vaccine virus is not likely to persist in tonsil tissue at such doses, the observed RNA-positivity is most likely due to a wild-type MV. However, further sequencing studies are needed to confirm the viral strain.

While the presence of MV in the intestinal tissue, lymph nodes and peripheral blood lymphocytes of subjects with no clinical measles has been discussed in the scientific literature, pharyngeal lymphoid tissues have not been investigated before.4,5 In their examination of nasopharyngeal secretion, Sonoda and Nakayama found no MV RNA, even in children whose blood lymphocytes were positive by PCR; they concluded that asymptomatic measles infection was common, but would rarely become a source of transmission because of negative PCR in the nasopahrynx.2 Our observation of MV RNA in tonsils suggests local retention of MV entering through the respiratory tract. At least two of these tonsil-positive cases had been immunized, implying a possible role of antibodies in this effect. The fact that serum antibody levels of these children were not higher than others might also support local, rather than systemic, antigenic stimulation.

These patients had chronic tonsillitis, and it may be speculated that the milieu of chronic inflammation impaired the clearance of MV and facilitated its latency. Because this hypothesis requires the examination of non-infected tonsil tissue from children, it is difficult to test.

The reason antigen and RNA do not always coexist in the same subject is explained by the low-level or focal tissue expression of the virus. Latent viruses do not always produce detectable viral proteins during latency; on the other hand, antigen may persist when there is no viral replication.

Because tissue samples had to be obtained from hospital cases and do not represent the general population, the present study was not designed to assess incidence. Nevertheless, the presence of MV RNA or antigen in people with no infectious symptoms or recent immunization supports the concept of MV latency in the normal population, especially if the population consists of both naturally infected and vaccinated subjects. It can be speculated that vaccinated individuals who possess serum IgG but not necessarily secretory IgA are likely to obtain local immunity by retaining MV in the pharyngeal lymphoid tissue upon exposure through the respiratory tract. The changing epidemiological features of measles in the post-vaccine era warrant the need for further studies to demonstrate differences between endemic and non-endemic regions.


  1. Top of page
  2. Acknowledgements
  3. References

The present study was supported by a grant from the Eczacibasi Scientific Research and Award Foundation and the Turkish Child Neurology Society.

We thank Professor S Ustacelebi for his review of the manuscript and helpful suggestions.


  1. Top of page
  2. Acknowledgements
  3. References
  • 1
    Katayama Y, Kohso K, Nishimura A, Tatsuno Y, Homma M, Hotta H. Detection of measles virus mRNA from autopsied human tissues. J. Clin. Microbiol. 1998; 36: 299301.
  • 2
    Sonoda S, Nakayama T. Detection of measles virus genome in lymphocytes from asymptomatic healthy children. J. Med. Virol. 2001; 65: 3817.
  • 3
    Kawashima H, Mori T, Takekuma K et al. Polymerase chain reaction detection of the hemagglutinin gene from an attenuated measles vaccine strain in the peripheral mononuclear cells of children with autoimmune hepatitis. Arch. Virol. 1996; 141: 87784.
  • 4
    Afzal MA, Armitage E, Ghosh S, Williams LC, Minor PD. Further evidence of the absence of measles virus genome sequence in full thickness intestinal specimens from patients with Crohn's disease. J. Med. Virol. 2000; 62: 37782.DOI: 10.1002/1096-9071(200011)62:3<377::aid-jmv10>;2-1
  • 5
    Kawashima H, Mori T, Kashiwagi Y, Takekuma K, Hoshika A, Wakefield A. Detection and sequencing of measles virus from peripheral mononuclear cells from patients with inflammatory bowel disease and autism. Dig. Dis. Sci.2000; 45: 7239.