SEARCH

SEARCH BY CITATION

Keywords:

  • HIV-1 infection;
  • Maternal antibodies;
  • Measles;
  • Measles immunization;
  • Neutralization test

Abstract

  1. Top of page
  2. Abstract
  3. Key notes
  4. Introduction
  5. Patients and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflicts of interest
  10. Source of funding
  11. References

Aim:  Growing numbers of newborns are saved from HIV infection through increased access to mother-to-child transmission prevention programmes. The maternally derived humoral immunity of these children might be impaired, both in terms of quantity and in terms of quality, with consequences for the timing of immunization against measles.

Methods:  A cell-ELISA technique compared the neutralizing activity on Edmonston strain measles virus of sera from 1- to 4-month-old infants. Ten serum specimens came from noninfected infants of HIV-infected mothers and another 10 from infants of healthy mothers. The sera were matched for the level of conventional ELISA measles antibodies.

Results:  Reflecting infection of the Vero cells by non-neutralized virus, optical density values were significantly higher for the sera from the children of the HIV-infected mothers than for those of the noninfected mothers (p < 0.001).

Conclusion:  Maternally derived protection against measles may be impaired by the mother’s HIV infection, relating to the quality rather than to the quantity of transplacental antibodies. Selective, early immunization with live attenuated measles vaccine should be evaluated in noninfected children of HIV-1-infected mothers.


Key notes

  1. Top of page
  2. Abstract
  3. Key notes
  4. Introduction
  5. Patients and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflicts of interest
  10. Source of funding
  11. References
  •  The constantly renewed pool of susceptible young infants, who lost their maternally derived antibodies at a varying age, is a challenge to any immunization programme and an obstacle to the eradication of measles.
  •  Noninfected babies of HIV-1 infected mothers may have weaker and shorter maternally derived protection due to the influence of the HIV infection on antibody production in the mother.
  •  These babies may need to be immunized against measles early.

Introduction

  1. Top of page
  2. Abstract
  3. Key notes
  4. Introduction
  5. Patients and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflicts of interest
  10. Source of funding
  11. References

With the eradication of measles high on the agenda of the WHO and other international organizations (1), it is important to define and target any new group of susceptible infants (2). Thanks to antiretroviral treatment (ART), the uninfected offspring of the HIV-infected mothers constitute a swiftly expanding group. In discussing protection against measles, these babies have largely been considered together with HIV-infected newborns. Still, there is reason to believe that even those saved from HIV infection by ART might be immunologically compromised, rendering them susceptible to measles and other infections, against which a healthy mother provides her full-term newborn with immunity via the placenta.

Earlier studies found relatively small (3) or conditional (high viral load) effects (4) of maternal HIV infection on placental transfer of antibodies to measles. Most of these studies used enzyme-linked immunosorbent assay (ELISA) for testing of specific immunoglobulin-G (IgG). With the plaque reduction neutralization assay, on the other hand, the majority of a group of 6-month-old infants born to HIV-infected mothers were estimated to be susceptible to measles and therefore immunizable with the live attenuated vaccine (5,6). This indicates that the standard immunization schedule with measles vaccine at 9 months may expose the noninfected children of HIV-infected mothers to unnecessary risk during a widened ‘age window of susceptibility’ to measles (7).

The classical neutralization test for measles antibodies is open to observer variation in assessing the cytopathic effect of the virus in a cell culture. Therefore, we used a new (‘cell-ELISA’) technique to measure the neutralizing activity of unknown sera on measles virus, comparing sera from non-HIV-infected infants born to HIV-infected and noninfected mothers.

Patients and methods

  1. Top of page
  2. Abstract
  3. Key notes
  4. Introduction
  5. Patients and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflicts of interest
  10. Source of funding
  11. References

Sample selection criteria

Index samples

Twenty-nine sera from 0- to 4-month-old infants born to HIV-1-infected mothers were examined for measles antibodies using a commercial ELISA (Enzygnost™). Ten sera from nine of these infants were selected for containing adequate levels of anti-measles antibodies and being sampled at similar ages as healthy control children with corresponding levels of ELISA antibodies.

Control samples

Sera from healthy infants 0–4 months old, previously examined for measles antibodies for the purpose of routine vaccination follow-up, were used as controls. Alike those from the children of the HIV-infected mothers, these sera were chosen for containing sizable concentrations of antibodies to measles as determined earlier by ELISA. The sera were paired on the basis of these antibody determinations, the age of each infant also being taken into account as far as possible.

Subjects

All the HIV-infected mothers were immigrants to Sweden and had acquired their immunity to measles and the HIV-1 infection elsewhere in the world. (See Table 1) They all received antiretroviral therapy and were monitored for viral load and CD4 cell count during pregnancy according to the Swedish mother-to-child transmission prevention programme (8). The infants were born between August 2008 and December 2009. They all tested negative with the Cobas TaqMan HIV-1 v2.0 (Roche Molecular Systems, Basel, Switzerland) at birth, 4–6 weeks and 6 months of age. The disappearance of the maternal antibodies was confirmed after 18 months of age with a combination assay for the detection of both HIV p24 antigen and antibodies against HIV-1/HIV-2 – Architect HIV Ag/Ab Combo (Abbott Diagnostics, Wiesbaden, Germany). See Table 1 for basic characteristics of these mothers and their infants. No serious morbidity was reported either in them or in the control infants. All of the latter were the result of pregnancies supervised according to Swedish standards, which include testing of the woman for HIV.

Table 1.   Characteristics of the examined infants and their HIV-infected mothers
MotherChild
AgeOriginParityCD4 cell count*Viral load (RNA copies/mL)*Gest. ageBirthSexBW (g)BL (cm)
  1. *Per mm3 of peripheral blood at gestational age about 30 weeks.

  2. PCS = Planned Caesarean section; ECS = Emergency Caesarean section.

  3. Samples from the same infant at different ages.

  4. §Examined at gestational age 37 weeks.

  5. Examined at gestational age 24 weeks.

28Ethiopia1G0P237<2037PCSM262048
26Ethiopia2G1P24086139ECSF317051
35Ethiopia2G1P3524037PCSF322650
25Ethiopia2G1P2526337PCSM270647
30Ivory Coast5G2P384<2036PCSF298049
26Ethiopia2G1P240861§39ECSF317051
33Uganda3G1P754<4037PCSM347049
33Kenya3G2P520887041VaginalM346550
35Thailand2G1P417<4032ECSF184643
31Ivory Coast1G0P513<4036PCSF237246

Measles antibody assessment

Cell-based neutralization assay

A novel technique was used for the neutralization test. It was inspired by previous work in the same laboratory, which, however, used immunofluorescence instead of enzymatic marking of infected cells (9). Extensive calibration was carried out in comparison with a conventional neutralization test, which is based on cytopathic effect observed in the microscope. It was found that the cell-ELISA was much more sensitive, and greater dilutions of the sera were needed with that technique.

The unknown sera were inactivated by heating in a waterbath to 56°C for 30 min and serially diluted in Eagle’s medium (with 2% foetal calf serum) in two times dilution steps, starting at 1/128, and aliquoted into 96-well plates. The pairs of sera that were going to be compared were placed on the same plate. Experimental controls were added to each plate in the form of (i) a reference serum (WHO standard) with known concentration of neutralizing antibodies to measles and (ii) only medium without serum or antibody.

A suspension of Edmonston strain measles virus was diluted 20 times in Eagle’s medium (as previously determined by titration). Fifty microlitres of this suspension was added to the equal volume of diluted serum in each well, and the plates were incubated at 37°C for one hour. Thereafter, the serum and virus suspension mixtures were transferred to 96-well plates, which had been prepared by letting cultured Vero cells adhere to the bottom of the wells as a monolayer. The plates were kept at 37°C in an incubator with 5% CO2 for 4 days. They were then washed gently three times with phosphate-buffered saline (PBS), and the cells in all the wells were fixed with 80% acetone at −20°C. Blocking of irrelevant antigenic sites was performed by adding 3% bovine serum albumin in PBS for 1 h at 37°C. Then, a monoclonal antibody to measles virus nucleocapsid antigen was added. After an hour at 37°C, the plates were carefully washed and a donkey anti-mouse alkaline phosphatase (ALP) conjugate (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) was added. After another hour of incubation, the diethanolamine substrate was added, together with levamisole solution to block any ALP activity of cellular origin. Optical density (OD) values were read in an ELISA reader at 405-nm after 40 min at room temperature.

The negative control wells had a mean OD value of 0.44 (range 0.402–0.477) with no rising tendency towards higher dilutions, whereas the unknown sera and the positive control all showed rising OD values (i.e. more antigen expressed on the Vero cells) with increasing dilution of the sera. These titration curves at some point made a leap of at least 0.2 from one dilution step to the next (at a level corresponding to that of the negative control). For the purpose of this study, the lower dilution of these two was defined as the titre of that sample, the interpretation being that it impeded infection of the Vero cells, whereas the higher of the two dilutions left some non-neutralized virus to infect the cells.

Measles antibody ELISA

The Enzygnost™ commercial assay kit (Dade Behring, Marburg, Germany) was used according to the manufacturer’s instructions. The specificity and sensitivity of the test are stated by the manufacturer as 99.6% and 100%, respectively. The plates were read in an ELISA reader at the 405-nm wavelength.

Ethical permission to use the sera from the infants of the HIV-infected mothers was obtained from the Regional Ethical Review Board in Stockholm (nos. 1998/98-184 and 2009/2035-32).

Statistical analysis

The Statistical Analysis System (SAS Institute, Frösunda, Stockholm, Sweden) under Linux was used for statistics.

Results

  1. Top of page
  2. Abstract
  3. Key notes
  4. Introduction
  5. Patients and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflicts of interest
  10. Source of funding
  11. References

The sera from the infants of the HIV-infected mothers (index children) were made available by the ongoing follow-up of the women with known HIV-1 infection in the Stockholm area, who participated in the European Collaborative Study on Mother-to-child transmission of HIV and gave birth between August 2008 and December 2009. The serum samples were selected that had been taken from these babies at ages where exposure to measles would be unlikely (<5 months). Each one of these sera was paired with one from a control child, based on the results of the Enzygnost ELISA. One of the infants contributed two blood specimens (marked in Table 1) drawn at different ages. Table 2 shows the slightly lower ELISA values of the index sera (mean = 1902; SD = 1229 and mean = 2095; SD = 1499, respectively) compared with the controls.

Table 2.   Anti-measles antibody activity by conventional ELISA (Enzygnost™) and by neutralization test* (NT) in the infants with HIV-1-infected mothers (index cases) and in the control infants
Index casesControls
Age (mos.)ELISA, (mU/mL)NT titreAge (mos.)ELISA (mU/mL)NT titre
  1. *cell-ELISA, see text for details.

  2. Lowest dilution step, from which optical density reading rose >0.2 to the next two times dilution of the serum.

1229125642685512
2173612841646128
1249012822533512
35512564458256
2201012822054128
48531283894256
1456925625700512
34811282757128
1272225612647512
31318128415781024

The neutralization assay was preliminarily interpreted in terms of a titre, as defined in Patients and Methods section. The positive control serum had the titre 256 or 512 on all the plates, and its Enzygnost reading was 1510 mU/mL. In the cell-based neutralization test, it produced a typical, rising curve through the dilution steps. None of the negative control wells exhibited a rise in the OD values of more than 0.1 between adjacent dilution steps. Table 2 shows that the index sera consistently had equal values to, or lower titre values than the controls.

To make a more detailed comparison of the neutralizing capacity of the sera from the two groups of infants, the OD values at the 256 times dilution were chosen (for being close to the titre of most of the specimens). The OD values were significantly higher for the sera from children to HIV-infected mothers than for the paired control sera (Fig. 1), indicating a lesser ability of the antibodies from the children of the infected mothers to neutralize the measles virus (p < 0.001 by paired t-test, df = 8, the repeat specimen from one child, and its paired control being left out of the calculation). Further, for the healthy control infants (crosses), Figure 1 shows graphically a negative linear association between the ELISA antibody values and the OD values in the neutralization test, while the values of the index children (rings) are more widely dispersed. As could also be expected, the WHO standard anti-measles serum gave a low OD value (i.e. high neutralizing activity) at a relatively high ELISA antibody concentration.

image

Figure 1.  Plot of optical density values in cell-ELISA neutralization test (vertical axis) vs. conventional solid-phase ELISA for measles antibodies (horizontal axis). Index serum specimens (exp) were from noninfected infants of HIV-1-infected mothers, and the crosses represent healthy control infants, matched to the index specimens for concentration of ELISA measles antibodies. Dot represents WHO standard anti-measles serum.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Key notes
  4. Introduction
  5. Patients and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflicts of interest
  10. Source of funding
  11. References

We found lower measles neutralizing activity in 10 sera from young infants born to HIV-infected mothers compared with matched controls with similar anti-measles activity values by conventional ELISA. While the difference in terms of measles antibody levels within the pairs could well be due to chance alone, the difference in OD values in the neutralization test was statistically significant (Fig. 1). Instead of using the conventional neutralization test, infection of cultured human cells was assessed with a ‘cell-ELISA’ approach. The advantages of this technique, in comparison with the classical neutralization tests, are the absence of subjective judgement in the evaluation of the results, the relatively small quantity of serum needed and the shorter incubations used in the assay. Drawbacks include the need for considerable technical skills, advanced laboratory equipment and the technique that cannot accommodate any great number of samples or great differences in titre values in the same assay round.

The novel method used in measuring measles virus neutralization was calibrated against the classical, but it requires considerably greater dilutions, so they could not be compared completely. Still, the above results make sense in at least two ways, providing an ‘internal’ validity check of sorts: all the specimens, and the positive control, gave gradually rising titration curves, containing a ‘leap’ at the level of the negative control (no neutralizing serum present), so that a critical dilution or titre could be determined. Further, for the specimens from the healthy control children, there was, just as could be expected, a graphically clear-cut negative association between high ELISA antibody titre and low OD in the neutralization test (the crosses in Fig. 1). Still, it remains to be found out whether this technique can serve generally as an alternative to the classical neutralization assays, for example to determine whether an individual is protected against measles.

The discrepancy between the results of the two antibody determination techniques supports the hypothesis of deficient humoral protection against measles in noninfected children to mothers with HIV-1 infection during pregnancy, indicating poorer quality rather than a difference in the quantity of the transferred IgG antibodies. De Moraes-Pinto et al. found reduced levels of IgG antibodies to measles in cord blood from women with HIV and placental malaria, compared with women without signs of these infections, whereas anti-tetanus antibody levels were unaffected (5). When comparing a large number of paired maternal cord sera from HIV-infected and noninfected mothers in Kenya, Scott et al. (10) found that children born to mothers with HIV were more likely to be seronegative to measles and to have lower antibody level. Farquhar et al. (4) found that 30 of 40 HIV-positive pregnant women had abnormally low anti-measles antibody titres and that reduction in placental transfer of IgG was associated with high HIV-1 viral load. In our study, the HIV-infected mothers were treated with combination ART and had undetectable or low viral load. However, the mentioned studies and others that found relatively limited effect of maternal HIV on transplacental immunity to measles did not test for neutralizing antibodies specifically. When the plaque reduction neutralization assay was applied to children of mothers with differing HIV status in Zambia, 86% of those who were born HIV-seropositive but uninfected were estimated to be susceptible to measles at 6 months (5). It would be interesting to include in the model of measles transmission in response to ART published by the same group, the possibility that noninfected children of HIV-1-infected mothers might lose their protection to measles at an early age.

The standard ELISA test measures both protective and nonprotective antibodies to many measles virus proteins, whereas tests for neutralization of in vitro infectivity should come closer to ‘real-life susceptibility’, either to naturally occurring measles or to the live attenuated vaccine. High-avidity antibodies are more likely to be protective, and in view of the negative consequences of HIV infection for the maturation of B cells (11), it would seem reasonable to believe that infected mothers would transfer less and more short-lived protection against measles to their offspring. The discrepancy between the results of the two test methods, as described above, would be predicted by that hypothesis.

Still, no in vitro test of neutralization will be a perfect model of real-life infectivity. The infecting measles virus in vitro is not the wild type, and the two probably have differential affinity for the cellular receptors that mediate virus entry into cells. Further, the cells used in the assay will differ from those of a live, susceptible person in many respects, including the expression of the relevant receptors. Nair et al. (12) penetrated these complications productively, comparing HIV-infected and uninfected children in Zambia. One of their conclusions was that several test methods need to be used in order to clarify protection and amenability to immunization against measles in children with HIV. Based on the above, we believe that the cited recommendation needs to be extended to the swiftly expanding group of children saved from the HIV infection by ART, and our study adds somewhat to existing knowledge in that sense. There is a need for field trials of immunization with live attenuated measles vaccine at ages below 9 months to uninfected children of mothers with HIV infection during pregnancy.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Key notes
  4. Introduction
  5. Patients and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflicts of interest
  10. Source of funding
  11. References

We thank the Swedish Institute for Communicable Disease Control (Smittskyddsinstitutet) and Dr Kari Johansen for generously providing laboratory facilities and Mrs Valentina Novak for skilful technical assistance.

References

  1. Top of page
  2. Abstract
  3. Key notes
  4. Introduction
  5. Patients and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflicts of interest
  10. Source of funding
  11. References
  • 1
    Strebel PM, Cochi SL, Hoekstra E, Rota PA, Featherstone D, Bellini WJ, et al. A world without measles. J Infect Dis 2011; 204(Suppl 1): S13.
  • 2
    WHO. Proceedings of the global technical consultation to assess the feasibility of measles eradication, 28–30 July 2010. J Infect Dis 2011; 204(Suppl 1): S413.
  • 3
    de Moraes-Pinto IM, Verhoeff F, Chimsuku L, Milligan PJM, Wesumperuma L, Broadhead RL, et al. Placental antibody transfer: influence of maternal HIV infection and placental malaria. Arch Dis Child Fetal Neonatal Ed 1998; 79: F2025.
  • 4
    Farquhar C, Nduati R, Haigwood N, Sutton W, Mbori-Ngacha D, Richardson B, et al. High maternal HIV-1 viral load during pregnancy is associated with reduced placental transfer of measles IgG antibody. J Acquir Immune Defic Syndr 2005; 40: 4947.
  • 5
    Scott S, Moss WJ, Cousens S, Beeler JA, Audet SA, Mugala N, et al. The influence of HIV-1 exposure and infection on levels of passively acquired antibodies to measles virus in Zambian infants. Clin Infect Dis 2007; 45: 141724.
  • 6
    Scott P, Moss WJ, Guilani Z, Low N. Measles vaccination in HIV-infected children: systematic review and meta-analysis of safety and immunogenicity. J Infect Dis 2011; 204(Suppl 1): S16478.
  • 7
    Feachem RG, Koblinsky MA. Interventions for the control of diarrhoeal diseases among young children: measles immunization. Bull World Health Organ 1983; 61: 64152.
  • 8
    Navér L, Bohlin A-B, Albert J, Flamholc L, Gisslén M, Gyllensten K, et al. Prophylaxis and treatment of HIV-1 infection in pregnancy: Swedish recommendations 2007. Scand J Infect Dis 2008; 40: 45161.
  • 9
    Vene S, Haglund M, Vapalahti O, Lundkvist Å. A rapid fluorescent focus inhibition test for detection of neutralizing antibodies to tick-borne encephalitis virus. J Virol Methods 1998; 73: 715.
  • 10
    Scott S, Cumberland P, Shulman CE, Cousens S, Cohen BJ, Brown DWG, et al. Neonatal measles immunity in rural Kenya: the influence of HIV and placental malaria infections on placental transfer of antibodies and levels of antibody in maternal and cord serum samples. J Infect Dis 2005; 191: 185460.
  • 11
    De Milito A, Nilsson A, Titanji K, Thorstensson R, Reizenstein E, Narita M, et al. Mechanisms of hypergammaglobulinemia and impaired antigen-specific humoral immunity in HIV-1 infection. Blood 2004; 103: 21806.
  • 12
    Nair N, Moss WJ, Scott S, Mugala N, Ndhlovu ZM, Lilo K, et al. HIV-1 infection in Zambian children impairs the development and avidity maturation of measles virus-specific immunoglobulin G after vaccination and infection. J Infect Dis 2009; 200: 10318.