Improved Detection of Latent Mycobacterium Tuberculosis Infection in HIV-1 Seropositive Individuals Using Cultured Cellular Assays.

Tuberculosis (TB), caused by the respiratory pathogen Mycobacterium tuberculosis (Mtb) is the leading cause of morbidity and mortality in patients coinfected with HIV-1. For the majority of immunocompetent individuals, Mtb replication is contained, and progression to active disease prevented by an orchestrated CD4+ T-cell specific and Th1-dominant immune response 1. These individuals are classified as latently infected with TB (LTBI) and have a 10% lifetime risk of developing TB 2. The progressive and generalized CD4+ T-cell depletion that occurs following HIV-1 infection increases the occurrence of TB disease in LTBI individuals to a 10% annual risk 2. Conversely, Mtb infection itself “aggravates” the course of HIV-1 infection, increasing viral loads and acting as a predictor for worsening clinical disease course 3. Current estimates are that one third of the world's population is infected with Mtb. Given that annual HIV-1 infection rates are increasing (http://apps.who.int/globalatlas/dataQuary/default.asp) 4, accurate diagnosis of both active and LTBI in HIV-1 positive individuals is of significant clinical value. Yet, to date, there is no gold standard for diagnosis of LTBI. Traditionally, the tuberculin skin test (TST) has been used to measure delayed hypersensitivity reaction following intra-dermal injection of purified protein derivative (PPD). However, the crude precipitate of Mtb culture supernatant (PPD) contains more than 200 proteins widely shared among mycobacteria, and consequently its use in the diagnosis of LTBI lacks both specificity and sensitivity 5, 6. More recently, IFN-γ release assays (IGRA) such as QuantiFERON-TB Gold In-Tube and T-SPOT.TB have been developed 7. IGRA improve the specificity of LTBI diagnosis by measuring cytokine production by immune cells to Mtb-specific proteins, transcribed from genes not found in BCG and most other environmental mycobacteria 7. Early secreted antigenic target-6 and culture filtrate-10 are the two best studied to date and can improve the sensitivity of LTBI diagnosis to 96% compared with 69% achieved by TST 7. Despite the advantages of using IGRA over TST, questions have been raised as to the usefulness of ex vivo IFN-γ-based T-cell assays in LTBI diagnosis in HIV-1 seropositive individuals 8, 9. HIV-1 infection is characterized by a reduction in the proliferative capacity and IFN-γ response of Mtb-specific CD4+ T cells. This, when coupled with the overall decline in CD4+ T-cell number and the poor concordance between different IGRA in immunocompromised individuals 8, 9, has raised concern as to the sensitivity of these assays in HIV-1 seropositive individuals 9-12, and prompted the Centre for Communicable Diseases (USA) to recommend caution in the use of IGRA when testing HIV-infected persons 13, 14. To address these issues, we sought to further improve the sensitivity of conventional ex vivo T.Spot, TB assays by specific expansion of TB-specific T cells in vitro. Cultured IFN-γ ELISpot has been shown to increase the sensitivity of detection of low frequency and low magnitude T-cell responses compared with standard ex vivo assays in a number of disease settings 15-17. The increased sensitivity is achieved by a 10-day, in vitro, peptide-stimulated culture, during which PBMC are supplemented with lymphoproliferative and anti-apoptotic cytokines (IL-2 and IL-7, respectively). Culture may enable expansion of antigen-specific T-cell populations that are both different from and additional to those detected by ex vivo assays to be measured 16. Using a laboratory-modified version of the T-Sport.TB test, we initially screened 27 HIV-1 seronegative and 27 seropositive Malawian individuals for ex vivo Mtb-specific T-cell responses. In accordance with current World Health Organisation (WHO) predictions and previous literature reports 4, 18, we found 13/27 HIV-1-negative individuals had Mtb-specific T-cell responses (48.1%) (Fig. 1A). On the contrary, HIV-1-positive participants had a much lower frequency, with only 3/27 HIV-1 participants (11.1%), having detectable Mtb-specific T-cell responses (Fig. 1B). We next compared the ex vivo responses detected in a subset of these participants (26 HIV-1 negative and 22 HIV-1 positive) with those observed following the cultured IFN-γ ELISpot technique. Briefly, 12/26 HIV-1 seronegative individuals had positive Mtb-specific T-cell responses detected ex vivo. Following cultured ELISpot, the magnitude of Mtb-specific responses increased >20fold, and, consistent with the previous studies 13-15, 17 the frequency at which Mtb-specific T-cell responses were detected among HIV-1 seronegative individuals increased from 12/26 ex vivo to 17/26 (65%) following cultured assay (Fig. 2A). More strikingly, a greater than 6-fold increase in the detection of Mtb-specific T-cell responses was observed in HIV-1 seropositive individuals tested; 1/22 ex vivo to 7/22 following culture (Fig. 2B). Importantly, Mtb-specific T-cell responses were detected in subjects with CD4 counts of <200 cells/mm3, suggesting that this technique overcomes some of the difficulties observed in applying IGRA to subjects with low CD4 counts 9, 12. Ex vivo IFN-γ ELISpot analysis of Mtb-specific responses in (A) 27 HIV-1 seronegative individuals and (B) 27 HIV-1-positive individuals from Malawi. Data presented as the number of spot forming units per million PBMC (SFU/106), following subtraction of mock-stimulated wells (background). All assays were performed in quadruplicate using Mtb-specific antigens (early secreted antigenic target-6 and culture filtrate-10), in addition to positive control antigens (PPD, FEC and PHA). Positive response defined as greater than or equal to 300 SFU/million PBMC, and or greater than 4× mock stimulation. Error bars represent standard error of mean. Mtb-specific IFN-γ responses detected using ex vivo ELISpot analysis (solid bars) were compared with responses detected following cultured IFN-γ ELISpot (hatched bars) in (A) HIV-1-negative and (B) HIV-positive individuals. All data presented as average number of SFU per million PBMC following background subtraction. Criteria for a positive response were 4× mock-stimulated well and greater than 50 SFU/million PBMC (ex vivo assays) or 300 SFU/million PBMC (cultured assay). Error bars shown represent standard error of mean. To examine whether the improved sensitivity resulted from nonspecific amplification or in vitro priming, cultured Mtb-specific T-cell responses were compared between a population in which Mtb infection is rare (UK – incidence 3%) and a population in which Mtb is endemic (Malawi – 68%) 4, 18. Following cultured ELISpot 26/41 (63%) HIV-1 seronegative, Malawian participants produced Mtb-specific T-cell responses (Supporting Information Fig. 1A), whereas 1/14 (7%) UK participants tested positive (Supporting Information Fig. 1B). Notably, the background for these assays was low and specific expansion of Mtb T cells was not observed in the absence of peptide stimulation (data not shown). The TB-specific response rates observed by cultured ELISpot are consistent with UK wide estimates of LTBI, suggesting that specificity was retained over the culture period of the assay. To our knowledge, this is the first demonstration that cultured IFN-γ ELISpot can be used to improve the detection of Mtb-specific T-cell responses among HIV-1-positive individuals. Notably, cultured PBMC can also be used for more detailed studies such as T-cell-epitope mapping and phenotyping using flow cytometry (data not shown). Although we have shown the cultured assay to be a reliable, experimental tool that is standardized for research purposes, it is not validated clinically. Additional studies are now required to determine the clinical relevance of cultured ELISpot responses and the LTBI rates in other cohorts from both TB endemic and nonendemic regions. However, the low cell number requirements mean that this technique should prove a valuable tool in future research of Mtb-specific T-cell responses in both HIV-1-negative and -positive individuals. In summary, these results confirm that current diagnostic techniques are underestimating latent Mtb infection rates and provide methodology for more accurate and detailed analysis of Mtb-specific responses in HIV-1/TB coinfection. The authors thank the study participants, the staff at the CHAVI repository CIS study team – CIS Study Team-Shannon Galvin, Clement Mapanje, Severiano Phakati, Maggrette Ndovi, Nancy Kamwendo, Mary Malahleka, Naomi Bonongwe, Naomi Nyirenda, Helen Milonde, Benard Kumwenda, Syze Gama, Dave Namakwa, Happiness Kanyamula, Beatrice Ndalama, Cecilia Massa, Julita Kenala, David Chilongozi, Francis Martinson, Irving Hoffman and Oxford laboratory members Victoria Bourne and Alasdair Leslie. This project is supported by: National Institute of Allergy and Infectious Diseases (NIAID) National Institutes of Health (NIH) Division of AIDS (DAIDS) US Department of Health and Human Services (HHS) Center for HIV/AIDS Vaccine Immunology (CHAVI) ♯ U19 AI067854-05. Conflict of interest: The authors declare no financial or commercial conflict of interest. Detailed facts of importance to specialist readers are published as ”Supporting Information”. Such documents are peer-reviewed, but not copy-edited or typeset. They are made available as submitted by the authors. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.