Search tips
Search criteria 


Logo of iaiPermissionsJournals.ASM.orgJournalIAI ArticleJournal InfoAuthorsReviewers
Infect Immun. 2017 April; 85(4): e01009-16.
Published online 2017 March 23. Prepublished online 2017 January 23. doi:  10.1128/IAI.01009-16
PMCID: PMC5364300

Characterization of T Cells Specific for CFP-10 and ESAT-6 in Mycobacterium tuberculosis-Infected Mauritian Cynomolgus Macaques

Sabine Ehrt, Editor
Sabine Ehrt, Weill Cornell Medical College;


Nonhuman primates can be used to study host immune responses to Mycobacterium tuberculosis. Mauritian cynomolgus macaques (MCMs) are a unique group of animals that have limited major histocompatibility complex (MHC) genetic diversity, such that MHC-identical animals can be infected with M. tuberculosis. Two MCMs homozygous for the relatively common M1 MHC haplotype were bronchoscopically infected with 41 CFU of the M. tuberculosis Erdman strain. Four other MCMs, which had at least one copy of the M1 MHC haplotype, were infected with a lower dose of 3 CFU M. tuberculosis. All animals mounted similar T-cell responses to CFP-10 and ESAT-6. Two epitopes in CFP-10 were characterized, and the MHC class II alleles restricting them were determined. A third epitope in CFP-10 was identified but exhibited promiscuous restriction. The CFP-10 and ESAT-6 antigenic regions targeted by T cells in MCMs were comparable to those seen in cases of human M. tuberculosis infection. Our data lay the foundation for generating tetrameric molecules to study epitope-specific CD4 T cells in M. tuberculosis-infected MCMs, which may guide future testing of tuberculosis vaccines in nonhuman primates.

KEYWORDS: CFP-10, ESAT-6, macaque, Mauritian cynomolgus macaque, T cells, T-cell immunity, tuberculosis, adaptive immunity, epitope, major histocompatibility complex


Tuberculosis (TB) is a significant threat to global health, and an effective vaccine to prevent infection with Mycobacterium tuberculosis or to contain M. tuberculosis replication is urgently needed. Animal models have played important roles in defining essential immune responses necessary for controlling M. tuberculosis infection and TB disease. T cells are one of the key immune cell types needed to contain M. tuberculosis within granulomas (1, 2), but relatively little characterization of these T cells has been performed.

Studying M. tuberculosis-specific T-cell responses within individual granulomas can be performed with samples from M. tuberculosis-infected animals but requires a detailed understanding of the T-cell populations within the host that are pathogen specific. Both rhesus macaques and cynomolgus macaques of Chinese origin historically have been used to study TB (3,6). Studies that map M. tuberculosis epitopes and define the restricting macaque major histocompatibility complex (MHC) alleles are limited (7, 8), but this information is a prerequisite for generating tetrameric reagents needed to enumerate these cells. Furthermore, those previous animal studies were complicated by the diversity of host MHC genetics and the promiscuous binding of M. tuberculosis-derived peptides by multiple MHC alleles (7).

Mauritian cynomolgus macaques (MCMs) are an increasingly common animal model in which to study T-cell responses to different pathogens (9). These animals have limited MHC genetic diversity, such that both MHC class I and II alleles have been characterized (10,12). Therefore, these animals provide a powerful tool to characterize the adaptive immune response to bacterial or viral infection. By infecting MHC-identical MCMs with M. tuberculosis, it is feasible to map epitopes in mycobacterial proteins and define the restricting MHC alleles for constructing tetrameric reagents, as others have done for simian immunodeficiency virus (SIV)-specific T cells (reviewed in reference 13). Developing reagents to effectively study M. tuberculosis-specific T cells in MCMs would improve this unique and powerful animal model and greatly benefit the TB pathogenesis and vaccine research community. For example, the ability to enumerate the number and breadth of antigen-specific T cells in granulomas that did or did not control M. tuberculosis infection might aid in the development of T-cell-based vaccines.

Until recently, no studies described infection of MCMs with M. tuberculosis. A recent study reported infection of MCMs with a high dose of M. tuberculosis via the aerosol route. TB in these animals rapidly progressed, and these animals were euthanized at 6 weeks postinfection (14). Such a rapid disease course complicates the characterization of epitope-specific T-cell responses. Since aerosolized M. tuberculosis infections of nonhuman primates (NHPs) may progress more rapidly than intrabronchial M. tuberculosis infections (5, 15), we infected MCMs by using a bronchoscope, a method capable of reproducibly and accurately delivering low-dose inocula that results in a spectrum of TB disease outcomes in Chinese cynomolgus macaques (3). We then assessed the development of M. tuberculosis-specific T cells that could be used in future studies of TB vaccine immunogenicity.


MCMs can be infected with M. tuberculosis and develop TB disease similar to those of humans and other macaques.

Two M1-homozygous MCMs (animals 36-15 and 37-15) were bronchoscopically infected with 41 CFU of the M. tuberculosis Erdman strain (Table 1). Serial positron emission tomography (PET)/computed tomography (CT) imaging using 2-deoxy-2-18F-deoxyglucose (18F-FDG) revealed steadily progressive TB disease in both animals, as indicated by abundant pulmonary lesions apparent on a CT scan, many of which were 18F-FDG avid, associated with higher metabolic activity and potentially indicative of an active host immune response (Fig. 1A and andB)B) (16). The M. tuberculosis inoculum was likely deposited on opposite sides in the two animals (Fig. 1A and andB);B); therefore, 18F-FDG-avid regions predominated in different lungs for the two animals. The humane endpoint was reached, and animals were euthanized at 93 days (animal 36-15) and 107 days (animal 37-15) postinfection. Four animals (animals 125-15, 126-15, 128-15, and 129-15) infected with a lower dose (3 CFU) of the M. tuberculosis Erdman strain exhibited a more diverse disease course. Two animals (animals 125-15 and 126-15) quickly developed advanced pulmonary disease characterized by numerous 18F-FDG-avid lesions, primarily in the left lung, and mediastinal lymphadenopathy (Fig. 1C and andD).D). These two animals reached the humane endpoint by 75 days (126-15) and 89 days (125-15) postinfection (Table 1). Also apparent were large areas of 18F-FDG uptake that were associated with large consolidations, as determined at necropsy in animals that presented with a more advanced disease course (Fig. 1A to toD).D). The other two animals (animals 128-15 and 129-15) exhibited more limited TB disease by PET/CT imaging that was relatively stably maintained for 5 months postinfection (Fig. 1E and andF),F), at which point the animals were euthanized (Table 1).

MHC genetics and M. tuberculosis infection doses for animals involved in this study
TB progression following M. tuberculosis infection of MCM. 18F-FDG PET/CT imaging reveals the extent of TB at 12 weeks (A, B, and D to F) or 11 weeks (C) postinfection with 41 CFU (animals 36-15 and 37-15) (A and B) or 3 CFU (animals 125-15, 126-15, 128-15, ...

ELISPOT analysis of MCMs infected with M. tuberculosis.

Whole blood was collected from the two animals infected with 41 CFU M. tuberculosis at 4, 6, and 8 weeks postinfection. At 4 weeks postinfection, peripheral blood mononuclear cells (PBMCs) from both animals were analyzed via a gamma interferon (IFN-γ) enzyme-linked immunosorbent spot (ELISPOT) assay for T-cell responses to pools of overlapping 15-mer peptides spanning the proteomes of the Ag85A, Ag85B, TB10.4, CFP-10, and ESAT-6 proteins of M. tuberculosis. The corresponding peptide regions found within each pool are indicated in Table 2. At 4 weeks postinfection, IFN-γ responses to peptide pools spanning all five of these proteins were detected (Fig. 2A). The most consistent responses in both animals were seen for peptide pools spanning the CFP-10 protein.

Corresponding amino acids of each M. tuberculosis-specific antigen contained within each peptide pool
Identification of antigenic regions in five M. tuberculosis proteins via ex vivo ELISPOT analysis of PBMCs from M. tuberculosis-infected MCMs. (A) At 4 weeks post-M. tuberculosis infection, PBMCs from infected MCMs were isolated and incubated overnight ...

To identify which specific peptides in CFP-10 and ESAT-6 were targeted by T cells, IFN-γ ELISPOT assays were performed with individual peptides comprising CFP-10 pools A, B, and C and ESAT-6 pool C by using PBMCs isolated from both animals at 6 weeks postinfection. These peptide pools were prioritized because T-cell responses to these antigens have been documented in cases of latent and active TB infection in humans as well as other NHP models (7, 17,20).

Responses specific for two 15-mer peptides in CFP-10 corresponding to amino acids 1 to 19 were detected in both animals (Fig. 2B). Even though the animals were MHC identical and had the potential to present the same peptides to T cells, the magnitude of the ELISPOT responses for both animals differed considerably for some peptides (e.g., CFP-1033–51) at the same time point (Fig. 2B). Weaker responses were detected for CFP-1057–71 and ESAT-665–83 in both animals (Fig. 2B).

Optimal peptides CFP-105–19 and CFP-1036–48 are presented by MHC class II DR and DQ molecules.

Peptide-specific polyclonal CD4+ T-cell lines were grown for the peptides eliciting positive ELISPOT responses. After ~4 to 5 weeks of in vitro culture, intracellular cytokine staining (ICS) assays were performed to determine whether the T-cell lines were specific for the indicated peptides. CD4+ T-cell lines specific for the CFP-101–19 and CFP-1033–51 peptides were successfully grown.

To determine the optimal epitope sequences, we tested two 15-mer overlapping peptides spanning each region and then smaller peptides within this region. We found that the region of CFP-10 spanning amino acids 5 to 19 yielded the most reproducible and robust cytokine responses at all dilutions tested for the CFP-101–19 region (Fig. 3A and andC).C). Additionally, the peptide spanning CFP-1036–48 yielded the most consistent response for the CFP-1033–51 region (Fig. 3B and andD).D). The differential production of cytokines observed with individual peptides of various lengths is likely attributed to the fact that MHC class II molecules have a more flexible binding pocket than do MHC class I molecules, allowing peptide length variability (21).

Mapping of optimal epitopes to CFP-101–19 and CFP-1033–51. (A and B) Polyclonal CD4+ T-cell lines were generated from CD8-depleted PBMCs to be specific for the CFP-101–19 (A) or CFP-1033–51 (B) antigenic region as described ...

Defining which MHC molecules restrict a given peptide is essential for constructing tetrameric reagents to track epitope-specific T cells. To determine this restriction, we generated MHC class II transferent cells by coexpressing the alpha and beta chains of the DR, DQ, and DP alleles present on the M1 MHC haplotype in RM3 cells. RM3 cells lack MHC class II expression (22), so only the transfected MHC class II molecules are presented. Each transferent cell line was pulsed with its respective optimal peptide and used as antigen-presenting cells in ICS assays with the epitope-specific cell lines. We found that the CFP-105–19 peptide was presented by Mafa-DQA1*2403/Mafa-DQB1*1801 molecules (Fig. 4A), and the CFP-1036–48 peptide was presented by Mafa-DRA*0201/Mafa-DRB*w501 molecules (Fig. 4B).

Restriction mapping of MHC class II responses to CFP-105–19 and CFP-1036–48. ICS assays were performed by using RM3 cells or RM3 cells stably transfected with the indicated MHC class II alleles associated with the M1 MCM haplotype to present ...

Epitope-specific CD4+ T cells are also elicited after infection with low-dose M. tuberculosis.

Even though epitope-specific CD4+ T cells were detected in MCMs infected with 41 CFU of M. tuberculosis, the rapid progression of TB disease (Fig. 1A and andB)B) implied that this dose might be too high for future studies that fully replicate the kinetics of an anti-M. tuberculosis T-cell response. In Chinese cynomolgus macaques, animals develop a spectrum of infection outcomes when a lower dose (~25 CFU) of M. tuberculosis Erdman is administered (3, 23), and the severity of TB disease tends to correlate with the infectious dose in rhesus macaques (24, 25). With the goal of using MCMs to study T-cell responses over the entire course of TB and for all disease outcomes, we sought to confirm that a lower dose of M. tuberculosis also elicited epitope-specific T cells.

We bronchoscopically infected four MCMs homozygous or heterozygous for the M1 MHC haplotype with just 3 CFU of M. tuberculosis. By 4 weeks postinfection, no significant IFN-γ ELISPOT responses were detectable in any animal (data not shown), which contrasted with the robust responses detected after infection with a high dose of 41 CFU of M. tuberculosis (Fig. 2). At 8 weeks postinfection, however, IFN-γ ELISPOT responses were detectable and specific for the same CFP-10 and ESAT-6 peptides that were observed in the animals infected with the high dose (Fig. 5). In the four animals infected with the lower dose of M. tuberculosis, we found that one animal (Fig. 5A, blue lines) responded to the CFP-105–19 epitope and that two of the four animals responded to the CFP-1036–48 epitope (Fig. 5A, blue lines), similar to the responses seen in the animals infected with the higher dose (Fig. 5A, red lines). However, despite sharing MHC genetics, the apparent magnitudes of the T-cell responses varied widely, even between animals infected with the same dose at the same time. Two of the animals infected with the lower M. tuberculosis dose (animals 126-15 and 125-15) produced little to no IFN-γ response to these CFP-10 regions (Fig. 5A, blue lines). We also examined the ELISPOT responses to 15-mer peptides spanning the ESAT-6 protein in all six animals (Fig. 5B). Four of the six animals, representing both high- and low-dose M. tuberculosis infection, showed responses to ESAT-665–83 (Fig. 5B).

Similar epitopes are targeted in MCMs infected with a high or low dose of M. tuberculosis (Mtb). Four M1-positive MCMs were infected with 3 CFU of the M. tuberculosis Erdman strain. At 8 weeks post-M. tuberculosis infection, PBMCs were isolated, and individual ...

The antigenic region of CFP-1071–85 is promiscuously restricted by two different M1 MHC class II molecules.

We also found a third region in CFP-10, CFP-1069–83, that was immunogenic in three of the four animals infected with a low dose of M. tuberculosis (Fig. 5A, blue lines). Immune responses to the same CFP-10 region were observed in animals infected with the high dose, but we were unable to grow CD4+ T-cell lines from them, so we could not determine the restricting MHC class II alleles. Using PBMCs from two animals infected with the low dose of M. tuberculosis (animals 128-15 and 129-15), we grew CFP-1071–85-specific T-cell lines that produced robust IFN-γ and/or tumor necrosis factor alpha (TNF-α) responses to a cognate peptide (Fig. 6A and andB).B). Unlike the other two epitopes that we mapped, CFP-1071–85 exhibited promiscuous MHC restriction. One T-cell line specific for CFP-1071–85 was restricted by Mafa-DQA*2403/Mafa-DQB*1801 (Fig. 6C), and the other was restricted by Mafa-DRA*0201/Mafa-DRB*w501 (Fig. 6D). Both molecules are present on the M1 MHC haplotype (11).

The CFP-1071–85 antigenic region is promiscuously restricted by M1 DR and DQ molecules. (A and B) Two CFP-1071–85-specific CD4 T-cell lines were generated from animal 128-15 (A) and animal 129-15 (B). ICS assays were conducted to determine ...

Regions of CFP-10 and ESAT-6 that are immunogenic in MCMs are similar to those found in humans.

We wanted to determine if the CFP-10 and ESAT-6 regions found to be immunogenic in MCMs were also immunogenic in humans. The immunogenic regions in CFP-10 and ESAT-6 that we identified in MCMs (Fig. 5) were compared to those reported to be immunogenic in humans (Tables 3 and and4).4). Table 3 summarizes the antigenic regions that we identified in MCMs: CFP-105–19, CFP-1036–48, CFP-1051–65, CFP-1071–85, and ESAT-665–83. Comparable antigenic regions were identified in individuals with latent and active TB (Table 4) (20, 26, 27). As we observed in MCMs, the CFP-1071–85 peptide region was promiscuously restricted by MHC class II DR and DQ molecules in humans (20). This suggests that T cells from MCMs and humans recognize similar portions of M. tuberculosis antigens, particularly for the CFP-10 protein.

Antigenic regions for MCM M. tuberculosis infection
Antigenic regions for human M. tuberculosis infection


We intrabronchially infected two MHC-identical M1-homozygous MCMs with 41 CFU of the M. tuberculosis Erdman strain. Both animals developed steadily progressive TB characterized by spreading and coalescing pulmonary granulomas, tuberculous pneumonia, and mediastinal lymphadenopathy, reaching humane endpoint criteria by 107 days postinfection. This TB disease course is consistent with that reported previously for MCMs following aerosol M. tuberculosis infection (14) and suggests that MCMs are relatively susceptible to M. tuberculosis. Since TB is often a more chronic disease, we wanted to characterize T-cell responses over a longer period of time. When we bronchoscopically infected four other MCMs with a much lower dose (3 CFU) of M. tuberculosis, two animals developed steadily progressive TB disease similar to that in animals that we infected with the higher dose. Notably, M. tuberculosis infection of the other two MCMs resulted in a much milder disease course, and the animals appeared clinically healthy and maintained relatively stable pulmonary disease as determined by monthly 18F-FDG PET/CT imaging. This suggests that when MCMs are bronchoscopically infected with a sufficiently low inoculum, they display a range of TB outcomes, similar to the spectrum of TB observed in cynomolgus macaques of Chinese origin (23).

We successfully mapped three epitopes in CFP-10 that were restricted by MHC class II molecules expressed in animals that have the M1 MHC haplotype. The restricting MHC class II molecules for two of the epitopes, CFP-105–19 and CFP-1036–48, were defined. Notably, the CFP-1071–85 epitope was promiscuously restricted by two different pairs of MHC class II molecules: Mafa-DQA*2403/DQB*1801 and Mafa-DRA*0201/DRB*w501. Presentation of the same epitope by multiple different MHC molecules has also been observed for human M. tuberculosis infections (20). This complex nuance of antigen presentation further supports the value of this animal model for studies of M. tuberculosis-specific T-cell immunity.

While we saw the most robust antigenic responses to CFP-10 in the ELISPOT analysis (Fig. 2A), we additionally saw ELISPOT responses to other antigens such as Ag85A, Ag85B, and TB10.4 (Fig. 2A). Due to the limited size of the animals and the amount of blood that could be safely drawn, as well as the well-documented immunogenic responses to CFP-10 and ESAT-6 in TB-infected humans and other nonhuman primates (7, 17, 18, 20), we chose to focus on the CFP-10 and ESAT-6 antigens for this study. However, due to the importance of Ag85A, TB10.4, and Ag85B in eliciting powerful immune responses and their recent incorporation into several vaccine studies (28,32), future studies could focus on the mapping of epitopes present in these antigens as well. Nonetheless, there are some TB diagnostic tools that use the CFP-10 and ESAT-6 proteins as immunogens (33, 34). If the regions of CFP-10 and ESAT-6 targeted by T cells in MCMs are similar to those targeted in humans, as our data suggest (Tables 3 and and4),4), MCMs would be an appropriate model to assess such assays and to determine how their results correspond to distinct TB disease states.

We observed that there were similar antigenic regions restricted by both human and MCM MHC molecules (Fig. 5 and Tables 3 and and4).4). This may be attributed, in part, to the sharing of allelic lineages between humans and MCMs at the DRB, DQA, and DQB loci (11). In addition to this, MHC class II has a very flexible binding pocket (21), which means that even MHC molecules that have some differences in their genetic sequences might still have the potential to present the same peptides. Due to this flexibility of the binding pocket, it was difficult to definitively map the optimal peptide that bound to MCM MHC molecules (Fig. 3). Overall, the similarity between human and macaque M. tuberculosis antigen presentations is intriguing and may strengthen the use of MCMs as a model system for human TB disease.

One unique advantage of the use of MCMs is that MHC-identical animals can be infected with the same strain and dose of a pathogen. In this way, non-MHC determinants of M. tuberculosis control can be studied. It is notable that the two MHC-identical animals (animals 36-15 and 37-15) exhibited similar disease progression post-M. tuberculosis infection. Nonetheless, the epitope-specific T-cell lines from one animal (animal 37-15) grew more readily than did those from the other animal (animal 36-15) (data not shown). We saw a striking disparity in TB progression in the four animals infected with a very low dose (3 CFU) of M. tuberculosis: two animals (animals 125-15 and 126-15) progressed to the humane endpoint less than 3 months after infection, and two (animals 128-15 and 129-15) controlled the disease for the length of the study (5 months postinfection). It is notable that both animals 126-15 and 129-15 were M1 homozygous and yet exhibited very different M. tuberculosis disease courses (Fig. 1D versus F and Table 1). Although the number of animals studied so far is too small to draw strong conclusions, this finding suggests that the MHC haplotype is not a major determinant of TB resistance, consistent with data from other studies (35). It is important to bear in mind that MHC-identical MCMs are not necessarily genetically identical outside the MHC, which may explain why our results contrast with the observation that monozygotic twin marmosets infected with the same strains of M. tuberculosis exhibited similar disease courses (36). In addition, the secretion of cytokines and presentation of antigens by macrophages, dendritic cells, or other innate cell types to initiate the adaptive immune response are well established to play a critical role in controlling M. tuberculosis replication and determining overall TB resistance (37,39). Ultimately, MHC-matched MCMs infected with the same dose and strain of M. tuberculosis may be valuable for defining non-MHC determinants of TB disease.

Besides the detection of T-cell responses in MCMs infected with a moderate dose (41 CFU) of M. tuberculosis, these same responses were detectable in animals infected with a very low M. tuberculosis dose (3 CFU). Furthermore, two of these animals infected with the low dose developed only minimal disease over the 5-month course of this study. These observations imply that a large disease burden is not required to elicit M. tuberculosis-specific T cells that are detectable in the circulation. Instead, a sufficient amount of antigen was present to elicit T cells targeting the same epitopes following a dose likely to be used for future TB studies in MCMs. We also found that peripheral T-cell responses developed more slowly in animals infected with the lower dose, as ELISPOT responses were apparent 2 to 4 weeks later in animals infected with 3 CFU M. tuberculosis, than in animals infected with 41 CFU M. tuberculosis. This was not surprising, as the kinetics of TB progression are directly proportional to the M. tuberculosis inoculum size in rhesus macaques (24, 25).

We have focused here on detecting T-cell responses in circulating blood. We are quite interested in characterizing the evolution of the anti-M. tuberculosis T-cell response at the site of infection, i.e., within individual granulomas, as responses in blood are not necessarily reflective of local responses (40). However, detecting peptide sequences is only one challenge when trying to characterize epitope-specific immune responses in lung tissue. It is very difficult to quantify the frequency of epitope-specific T cells in individual granulomas from NHPs without the use of tetrameric reagents. Even though we identified just three M. tuberculosis epitopes in this study, we importantly identified the restricting MHC class II molecules using MHC class II transferent cell lines. Since we now know the immunogenic M. tuberculosis peptides as well as their corresponding restricting MHC class II alleles, we can construct tetrameric reagents to enumerate these T cells in tissues. Notably, animals with M1 MHC class II alleles are present at a frequency of 31% across the MCM population (11) and so should be readily available for such studies.

Together, we show that MCMs can be infected intrabronchially with a moderate dose (41 CFU) and a very low dose (3 CFU) of the pathogenic M. tuberculosis Erdman strain. Even with this higher dose of M. tuberculosis, MCMs survived at least 12 weeks, whereas aerosol-infected MCMs survived only ~6 weeks (14). We were able to detect peripheral T-cell responses to M. tuberculosis in all animals. Thus, MCMs represent a novel and viable animal model in which to study T-cell responses to M. tuberculosis. Since T-cell responses are essential for effective host anti-M. tuberculosis immunity (1, 2), the data that we describe here will be instrumental in generating immunologic tools for future studies of T-cell responses to M. tuberculosis. Furthermore, these data validate the use of MHC-identical MCMs infected with M. tuberculosis in order to characterize epitope-specific T cells in response to mycobacterial proteins.


Animals and M. tuberculosis infection.

All six animals used in this study were obtained from Bioculture, Ltd. (Mauritius); quarantined at Buckshire (Perkasie, PA); and then housed in a biosafety level 3 facility at the University of Pittsburgh for M. tuberculosis infection. Two MCMs were infected bronchoscopically, as previously described (3), with 41 CFU of M. tuberculosis (Erdman strain), and four MCMs were infected with 3 CFU (Table 1). Aliquots of each inoculum were plated onto 7H11 agar and incubated at 37°C for 3 weeks to confirm the infectious dose. Blood was collected under ketamine sedation via inguinal venipuncture using EDTA Vacutainer tubes. At monthly intervals postinfection, animals were sedated, mechanically ventilated, and imaged by 2-deoxy-2-18F-deoxyglucose (18F-FDG) PET and CT using a microPET Focus 220 preclinical PET scanner (Siemens Molecular Solutions, Knoxville, TN) coupled to a Neurologica CT scanner (Neurologica Corp., Danvers, MA) as previously described (16). All animal protocols and procedures were approved by the University of Pittsburgh's Institutional Animal Care and Use Committee, which adheres to guidelines established by the Animal Welfare Act and the Guide for the Care and Use of Laboratory Animals (41).


All peptides of 15 amino acids in length were obtained through BEI Resources (Manassas, VA). These peptides consisted of 15-mer peptides (overlapping by 11 amino acids) spanning the entire amino acid sequences for the ESAT-6, CFP-10, Ag85A, Ag85B, and TB10.4 proteins of M. tuberculosis. Subsequent peptides of various lengths used for the mapping of optimal epitope sequences were synthesized by GenScript Services (Piscataway, NJ).

Cell lines and clones.

Polyclonal CD8+ and CD4+ T-cell lines were generated from whole-blood PBMCs that were separated by Ficoll density gradient separation. CD8+ T-cell lines were generated as previously described (10, 42). To generate CD4+ T-cell lines, PBMCs were depleted of CD8+ cells by using anti-CD8 microbeads (specific for NHPs) via magnetic separation according to the manufacturer's protocol (Miltenyi Biotech, San Diego, CA). The depletion of CD8+ cells was confirmed by surface staining with anti-CD3 Alexa Fluor 700 (AF700), anti-CD8 PacBlue (BD Biosciences, San Jose, CA), and anti-CD4 allophycocyanin (APC) (Miltenyi Biotech) antibodies, which depleted ~95 to 98% of all CD8+ cells (data not shown). The CD8-depleted cells were then grown in RPMI medium (Thermo Fisher Scientific, Grand Island, NY) containing 15% fetal bovine serum (FBS; Corning, Tewksbury, MA), 4 mM l-glutamine (Thermo Fisher Scientific), 1% antibiotic-antimycotic agent (Thermo Fisher Scientific), and 50 ng/ml interleukin-7 (IL-7) (BioLegend, San Diego, CA) for 1 week in the presence of specific M. tuberculosis peptides. After 1 week, medium was replaced with identical medium except that IL-7 was replaced with 100 U/ml IL-2 (Prometheus Laboratories, San Diego, CA). CD8-depleted, CD4+ T-cell lines were cultured as previously described (43) in the presence of autologous B-lymphocytic cell lines (BLCLs) (10, 42) pulsed with M. tuberculosis-specific peptides.

Intracellular cytokine staining assay.

To determine the specificity of the CD4+ T-cell lines and identify the restricting MHC class II molecules, ICS was used as previously described (42, 44). Briefly, 2 × 105 autologous BLCLs or RM3-MHC transferent cells (see below) were pulsed with 1 μM peptide for 90 min at 37°C with 5% CO2. As a negative control, cells were pulsed with no peptide. After peptide pulsing, cells were washed twice with RPMI medium containing 10% FBS (R10) and then resuspended in 100 μl of R10. A total of 2 × 105 cells of the CD4+ T-cell line was then added to each tube, along with brefeldin A (10-μg/ml final concentration; Sigma-Aldrich, St. Louis, MO) to halt the Golgi-mediated transport of proteins. After incubation at 37°C with 5% CO2 for ~5 h, the cells were stained with anti-CD3 AF700, anti-CD8 PacBlue, and anti-CD4 APC for 20 min at room temperature in the dark; washed twice with phosphate-buffered saline containing 10% FBS (fluorescence-activated cell sorter [FACS] buffer); and fixed with 2% paraformaldehyde (PFA) overnight. The following day, cells were permeabilized with 0.1% saponin and stained with anti-IFN-γ fluorescein isothiocyanate (FITC) (BD biosciences) and anti-TNF-α peridinin chlorophyll protein (PerCP) Cy5.5 (BioLegend) antibodies for 20 min at room temperature in the dark. Cells were washed twice with saponin buffer and once with FACS buffer and then fixed with 2% PFA and stored until analysis. Flow cytometry was performed on a BD LSR II instrument (Becton Dickinson, Franklin Lakes, NJ), and data were analyzed by using FlowJo software for Macintosh (version 9.9.3).

ELISPOT assay.

An IFN-γ ELISPOT assay was used, as previously described (42, 45), to determine the peptide specificity of CD4+ and CD8+ T cells isolated from M. tuberculosis-infected animals. Briefly, peptide pools were generated by combining 7 to 8 individual 15-mer peptides in equal volumes (10 mM stock concentration, overlapping by 11 amino acids). These pools represented the entire proteome for M. tuberculosis ESAT-6, CFP-10, Ag85A, Ag85B, and TB10.4 proteins. Freshly isolated PBMCs (1 × 105) were then added to each well of an NHP IFN-γ ELISPOT plate (Mabtech, Sweden) with either 2.5 μM the indicated peptide pools or 1 μM the individual 15-mer peptides. As a negative control, 1 × 105 cells were incubated with medium alone; as a positive control, 1 × 105 cells were incubated with 10 μg/ml concanavalin A. The plates were incubated overnight at 37°C and 5% CO2 and were developed the following day according to the manufacturer's protocol, as previously described (42, 45). Plates were read by using an AID robotic ELISPOT reader (AID, Strassberg, Germany). A positive response was defined as the number of spot-forming colonies (SFCs) per 106 PBMCs that was 2 standard deviations above the average value for the negative control or 50 SFCs/106 PBMCs, whichever was greater.

RM3 transferents.

The RM3 cell line is a derivative of the B-lymphocytic Raji cell line that lacks the expression of MHC class II molecules (22). Stable transferents expressing the alpha and beta alleles of the MCM M1 DR, DQ, and DP gene loci were constructed as follows: the pCEP4-Mafa-DRA*0103, pCEP4-Mafa-DRB*w2101, and pCEP4-Mafa-DRB*w501 plasmid vectors were obtained from Jonah Sacha (Oregon Health and Science University, Portland, OR) and were constructed as previously described (43). To create plasmid vectors for the MCM M1 DQ and DP gene alleles, the alpha and beta gene pairs (Table 5) were PCR amplified from cDNA and cloned into the pCEP4-hygromycin vector (Thermo Fisher Scientific) by using HindIII and NotI restriction sites. The vectors listed above and in Table 5 were sequence verified prior to use. The alpha and beta gene pairs for each DR, DQ, and DP allele were then electroporated into RM3 cells by using the Neon transfection system 10-μl kit according to the manufacturer's instructions (Thermo Fisher Scientific). Stable RM3 transferents were grown in R10 supplemented with 400 μg/ml hygromycin B (Thermo Fisher Scientific). After 4 to 6 weeks of hygromycin selection, the expression of the desired MHC class II allele was tested by staining for 20 min with either anti-HLA DR AF700 (BD Biosciences) or anti-HLA DR, DP, or DQ FITC (BD Biosciences). The transferents were fixed with 2% PFA. Flow cytometry was performed as described above.

MHC class II α and β gene pairs for the MCM MHC M1 haplotype


We thank Jonah Sacha (Oregon Health and Science University) for kindly providing the pCEP4 MHC class II plasmids and JoAnne Flynn (University of Pittsburgh) for helpful advice. We also thank members of David O'Connor's laboratory for experimental advice, members of the Wisconsin National Primate Center Research Services unit for MHC genotyping of animals, and members of JoAnne Flynn's laboratory for invaluable assistance with the MCM studies.

These studies were supported by NIH grant RO1 AI-111815 and by a University of Pittsburgh School of Medicine Dean's award granted to C.A.S. This research was conducted in part at a facility constructed with support from Research Facilities Improvement Program grants RR15459-01 and RR020141-01. We also thank members of the Wisconsin National Primate Research Center, a facility supported by grants P51RR000167 and P51OD011106. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.

We declare no competing interests.


1. Cooper AM, Flynn JL 1995. The protective immune response to Mycobacterium tuberculosis. Curr Opin Immunol 7:512–516. doi:.10.1016/0952-7915(95)80096-4 [PubMed] [Cross Ref]
2. Flynn JL, Chan J 2001. Immunology of tuberculosis. Annu Rev Immunol 19:93–129. doi:.10.1146/annurev.immunol.19.1.93 [PubMed] [Cross Ref]
3. Capuano SV, Croix DA, Pawar S, Zinovik A, Myers A, Lin PL, Bissel S, Fuhrman C, Klein E, Flynn JL 2003. Experimental Mycobacterium tuberculosis infection of cynomolgus macaques closely resembles the various manifestations of human M. tuberculosis infection. Infect Immun 71:5831–5844. doi:.10.1128/IAI.71.10.5831-5844.2003 [PMC free article] [PubMed] [Cross Ref]
4. Flynn JL, Capuano SV, Croix D, Pawar S, Myers A, Zinovik A, Klein E 2003. Non-human primates: a model for tuberculosis research. Tuberculosis (Edinb) 83:116–118. doi:.10.1016/S1472-9792(02)00059-8 [PubMed] [Cross Ref]
5. Scanga CA, Flynn JL 2014. Modeling tuberculosis in nonhuman primates. Cold Spring Harb Perspect Med 4:a018564. doi:.10.1101/cshperspect.a018564 [PMC free article] [PubMed] [Cross Ref]
6. Lewinsohn DM, Tydeman IS, Frieder M, Grotzke JE, Lines RA, Ahmed S, Prongay KD, Primack SL, Colgin LM, Lewis AD, Lewinsohn DA 2006. High resolution radiographic and fine immunologic definition of TB disease progression in the rhesus macaque. Microbes Infect 8:2587–2598. doi:.10.1016/j.micinf.2006.07.007 [PubMed] [Cross Ref]
7. Mothé BR, Lindestam Arlehamn CS, Dow C, Dillon MB, Wiseman RW, Bohn P, Karl J, Golden NA, Gilpin T, Foreman TW, Rodgers MA, Mehra S, Scriba TJ, Flynn JL, Kaushal D, O'Connor DH, Sette A 2015. The TB-specific CD4(+) T cell immune repertoire in both cynomolgus and rhesus macaques largely overlap with humans. Tuberculosis (Edinb) 95:722–735. doi:.10.1016/ [PMC free article] [PubMed] [Cross Ref]
8. Wei H, Wang R, Yuan Z, Chen CY, Huang D, Halliday L, Zhong W, Zeng G, Shen Y, Shen L, Wang Y, Chen ZW 2009. DR*W201/P65 tetramer visualization of epitope-specific CD4 T-cell during M. tuberculosis infection and its resting memory pool after BCG vaccination. PLoS One 4:e6905. doi:.10.1371/journal.pone.0006905 [PMC free article] [PubMed] [Cross Ref]
9. Wiseman RW, Karl JA, Bohn PS, Nimityongskul FA, Starrett GJ, O'Connor DH 2013. Haplessly hoping: macaque major histocompatibility complex made easy. ILAR J 54:196–210. doi:.10.1093/ilar/ilt036 [PMC free article] [PubMed] [Cross Ref]
10. Budde ML, Lhost JJ, Burwitz BJ, Becker EA, Burns CM, O'Connor SL, Karl JA, Wiseman RW, Bimber BN, Zhang GL, Hildebrand W, Brusic V, O'Connor DH 2011. Transcriptionally abundant major histocompatibility complex class I alleles are fundamental to nonhuman primate simian immunodeficiency virus-specific CD8+ T cell responses. J Virol 85:3250–3261. doi:.10.1128/JVI.02355-10 [PMC free article] [PubMed] [Cross Ref]
11. O'Connor SL, Blasky AJ, Pendley CJ, Becker EA, Wiseman RW, Karl JA, Hughes AL, O'Connor DH 2007. Comprehensive characterization of MHC class II haplotypes in Mauritian cynomolgus macaques. Immunogenetics 59:449–462. doi:.10.1007/s00251-007-0209-7 [PMC free article] [PubMed] [Cross Ref]
12. Wiseman RW, Wojcechowskyj JA, Greene JM, Blasky AJ, Gopon T, Soma T, Friedrich TC, O'Connor SL, O'Connor DH 2007. Simian immunodeficiency virus SIVmac239 infection of major histocompatibility complex-identical cynomolgus macaques from Mauritius. J Virol 81:349–361. doi:.10.1128/JVI.01841-06 [PMC free article] [PubMed] [Cross Ref]
13. Valentine LE, Watkins DI 2008. Relevance of studying T cell responses in SIV-infected rhesus macaques. Trends Microbiol 16:605–611. doi:.10.1016/j.tim.2008.08.010 [PMC free article] [PubMed] [Cross Ref]
14. Javed S, Marsay L, Wareham A, Lewandowski KS, Williams A, Dennis MJ, Sharpe S, Vipond R, Silman N, Ball G, Kempsell KE 2016. Correction: temporal expression of peripheral blood leukocyte biomarkers in a Macaca fascicularis infection model of tuberculosis; comparison with human datasets and analysis with parametric/non-parametric tools for improved diagnostic biomarker identification. PLoS One 11:e0159783. doi:.10.1371/journal.pone.0159783 [PMC free article] [PubMed] [Cross Ref]
15. Sharpe SA, Eschelbach E, Basaraba RJ, Gleeson F, Hall GA, McIntyre A, Williams A, Kraft SL, Clark S, Gooch K, Hatch G, Orme IM, Marsh PD, Dennis MJ 2009. Determination of lesion volume by MRI and stereology in a macaque model of tuberculosis. Tuberculosis (Edinb) 89:405–416. doi:.10.1016/ [PubMed] [Cross Ref]
16. Lin PL, Coleman T, Carney JP, Lopresti BJ, Tomko J, Fillmore D, Dartois V, Scanga C, Frye LJ, Janssen C, Klein E, Barry CE III, Flynn JL 2013. Radiologic responses in cynomolgus macaques for assessing tuberculosis chemotherapy regimens. Antimicrob Agents Chemother 57:4237–4244. doi:.10.1128/AAC.00277-13 [PMC free article] [PubMed] [Cross Ref]
17. Brodin P, Rosenkrands I, Andersen P, Cole ST, Brosch R 2004. ESAT-6 proteins: protective antigens and virulence factors. Trends Microbiol 12:500–508. doi:.10.1016/j.tim.2004.09.007 [PubMed] [Cross Ref]
18. Skjøt RL, Oettinger T, Rosenkrands I, Ravn P, Brock I, Jacobsen S, Andersen P 2000. Comparative evaluation of low-molecular-mass proteins from Mycobacterium tuberculosis identifies members of the ESAT-6 family as immunodominant T-cell antigens. Infect Immun 68:214–220. doi:.10.1128/IAI.68.1.214-220.2000 [PMC free article] [PubMed] [Cross Ref]
19. Andersen P, Munk ME, Pollock JM, Doherty TM 2000. Specific immune-based diagnosis of tuberculosis. Lancet 356:1099–1104. doi:.10.1016/S0140-6736(00)02742-2 [PubMed] [Cross Ref]
20. Arlehamn CS, Sidney J, Henderson R, Greenbaum JA, James EA, Moutaftsi M, Coler R, McKinney DM, Park D, Taplitz R, Kwok WW, Grey H, Peters B, Sette A 2012. Dissecting mechanisms of immunodominance to the common tuberculosis antigens ESAT-6, CFP10, Rv2031c (hspX), Rv2654c (TB77), and Rv1038c (EsxJ). J Immunol 188:5020–5031. doi:.10.4049/jimmunol.1103556 [PMC free article] [PubMed] [Cross Ref]
21. Ferrante A. 2013. For many but not for all: how the conformational flexibility of the peptide/MHCII complex shapes epitope selection. Immunol Res 56:85–95. doi:.10.1007/s12026-012-8342-2 [PMC free article] [PubMed] [Cross Ref]
22. Calman AF, Peterlin BM 1987. Mutant human B cell lines deficient in class II major histocompatibility complex transcription. J Immunol 139:2489–2495. [PubMed]
23. Lin PL, Rodgers M, Smith L, Bigbee M, Myers A, Bigbee C, Chiosea I, Capuano SV, Fuhrman C, Klein E, Flynn JL 2009. Quantitative comparison of active and latent tuberculosis in the cynomolgus macaque model. Infect Immun 77:4631–4642. doi:.10.1128/IAI.00592-09 [PMC free article] [PubMed] [Cross Ref]
24. Zhang J, Xian Q, Guo M, Huang Z, Rao Y, Wang Y, Wang X, Bao R, Evans TG, Hokey D, Sizemore D, Ho WZ 2014. Mycobacterium tuberculosis Erdman infection of rhesus macaques of Chinese origin. Tuberculosis (Edinb) 94:634–643. doi:.10.1016/ [PubMed] [Cross Ref]
25. Zhang J, Ye YQ, Wang Y, Mo PZ, Xian QY, Rao Y, Bao R, Dai M, Liu JY, Guo M, Wang X, Huang ZX, Sun LH, Tang ZJ, Ho WZ 2011. M. tuberculosis H37Rv infection of Chinese rhesus macaques. J Neuroimmune Pharmacol 6:362–370. doi:.10.1007/s11481-010-9245-4 [PubMed] [Cross Ref]
26. Paul S, Dillon MB, Lindestam Arlehamn CS, Huang H, Davis MM, McKinney DM, Scriba TJ, Sidney J, Peters B, Sette A 2015. A population response analysis approach to assign class II HLA-epitope restrictions. J Immunol 194:6164–6176. doi:.10.4049/jimmunol.1403074 [PMC free article] [PubMed] [Cross Ref]
27. Lewinsohn DA, Winata E, Swarbrick GM, Tanner KE, Cook MS, Null MD, Cansler ME, Sette A, Sidney J, Lewinsohn DM 2007. Immunodominant tuberculosis CD8 antigens preferentially restricted by HLA-B. PLoS Pathog 3:e127. doi:.10.1371/journal.ppat.0030127 [PMC free article] [PubMed] [Cross Ref]
28. Scriba TJ, Tameris M, Mansoor N, Smit E, van der Merwe L, Isaacs F, Keyser A, Moyo S, Brittain N, Lawrie A, Gelderbloem S, Veldsman A, Hatherill M, Hawkridge A, Hill AV, Hussey GD, Mahomed H, McShane H, Hanekom WA 2010. Modified vaccinia Ankara-expressing Ag85A, a novel tuberculosis vaccine, is safe in adolescents and children, and induces polyfunctional CD4+ T cells. Eur J Immunol 40:279–290. doi:.10.1002/eji.200939754 [PMC free article] [PubMed] [Cross Ref]
29. Radosevic K, Wieland CW, Rodriguez A, Weverling GJ, Mintardjo R, Gillissen G, Vogels R, Skeiky YA, Hone DM, Sadoff JC, van der Poll T, Havenga M, Goudsmit J 2007. Protective immune responses to a recombinant adenovirus type 35 tuberculosis vaccine in two mouse strains: CD4 and CD8 T-cell epitope mapping and role of gamma interferon. Infect Immun 75:4105–4115. doi:.10.1128/IAI.00004-07 [PMC free article] [PubMed] [Cross Ref]
30. McShane H, Pathan AA, Sander CR, Goonetilleke NP, Fletcher HA, Hill AV 2005. Boosting BCG with MVA85A: the first candidate subunit vaccine for tuberculosis in clinical trials. Tuberculosis (Edinb) 85:47–52. doi:.10.1016/ [PubMed] [Cross Ref]
31. Pérez de Val B, Villarreal-Ramos B, Nofrarías M, López-Soria S, Romera N, Singh M, Abad FX, Xing Z, Vordermeier HM, Domingo M 2012. Goats primed with Mycobacterium bovis BCG and boosted with a recombinant adenovirus expressing Ag85A show enhanced protection against tuberculosis. Clin Vaccine Immunol 19:1339–1347. doi:.10.1128/CVI.00275-12 [PMC free article] [PubMed] [Cross Ref]
32. Xing Z, McFarland CT, Sallenave JM, Izzo A, Wang J, McMurray DN 2009. Intranasal mucosal boosting with an adenovirus-vectored vaccine markedly enhances the protection of BCG-primed guinea pigs against pulmonary tuberculosis. PLoS One 4:e5856. doi:.10.1371/journal.pone.0005856 [PMC free article] [PubMed] [Cross Ref]
33. Ganguly N, Siddiqui I, Sharma P 2008. Role of M. tuberculosis RD-1 region encoded secretory proteins in protective response and virulence. Tuberculosis (Edinb) 88:510–517. doi:.10.1016/ [PubMed] [Cross Ref]
34. Pai M, Denkinger CM, Kik SV, Rangaka MX, Zwerling A, Oxlade O, Metcalfe JZ, Cattamanchi A, Dowdy DW, Dheda K, Banaei N 2014. Gamma interferon release assays for detection of Mycobacterium tuberculosis infection. Clin Microbiol Rev 27:3–20. doi:.10.1128/CMR.00034-13 [PMC free article] [PubMed] [Cross Ref]
35. Hill AV. 2006. Aspects of genetic susceptibility to human infectious diseases. Annu Rev Genet 40:469–486. doi:.10.1146/annurev.genet.40.110405.090546 [PubMed] [Cross Ref]
36. Via LE, Weiner DM, Schimel D, Lin PL, Dayao E, Tankersley SL, Cai Y, Coleman MT, Tomko J, Paripati P, Orandle M, Kastenmayer RJ, Tartakovsky M, Rosenthal A, Portevin D, Eum SY, Lahouar S, Gagneux S, Young DB, Flynn JL, Barry CE 2013. Differential virulence and disease progression following Mycobacterium tuberculosis complex infection of the common marmoset (Callithrix jacchus). Infect Immun 81:2909–2919. doi:.10.1128/IAI.00632-13 [PMC free article] [PubMed] [Cross Ref]
37. Sia JK, Georgieva M, Rengarajan J 2015. Innate immune defenses in human tuberculosis: an overview of the interactions between Mycobacterium tuberculosis and innate immune cells. J Immunol Res 2015:747543. doi:.10.1155/2015/747543 [PMC free article] [PubMed] [Cross Ref]
38. Lerner TR, Borel S, Gutierrez MG 2015. The innate immune response in human tuberculosis. Cell Microbiol 17:1277–1285. doi:.10.1111/cmi.12480 [PMC free article] [PubMed] [Cross Ref]
39. van Crevel R, Ottenhoff TH, van der Meer JW 2002. Innate immunity to Mycobacterium tuberculosis. Clin Microbiol Rev 15:294–309. doi:.10.1128/CMR.15.2.294-309.2002 [PMC free article] [PubMed] [Cross Ref]
40. Gideon HP, Phuah J, Myers AJ, Bryson BD, Rodgers MA, Coleman MT, Maiello P, Rutledge T, Marino S, Fortune SM, Kirschner DE, Lin PL, Flynn JL 2015. Variability in tuberculosis granuloma T cell responses exists, but a balance of pro- and anti-inflammatory cytokines is associated with sterilization. PLoS Pathog 11:e1004603. doi:.10.1371/journal.ppat.1004603 [PMC free article] [PubMed] [Cross Ref]
41. National Research Council. 2011. Guide for the care and use of laboratory animals, 8th ed National Academies Press, Washington, DC.
42. Burwitz BJ, Pendley CJ, Greene JM, Detmer AM, Lhost JJ, Karl JA, Piaskowski SM, Rudersdorf RA, Wallace LT, Bimber BN, Loffredo JT, Cox DG, Bardet W, Hildebrand W, Wiseman RW, O'Connor SL, O'Connor DH 2009. Mauritian cynomolgus macaques share two exceptionally common major histocompatibility complex class I alleles that restrict simian immunodeficiency virus-specific CD8+ T cells. J Virol 83:6011–6019. doi:.10.1128/JVI.00199-09 [PMC free article] [PubMed] [Cross Ref]
43. Giraldo-Vela JP, Rudersdorf R, Chung C, Qi Y, Wallace LT, Bimber B, Borchardt GJ, Fisk DL, Glidden CE, Loffredo JT, Piaskowski SM, Furlott JR, Morales-Martinez JP, Wilson NA, Rehrauer WM, Lifson JD, Carrington M, Watkins DI 2008. The major histocompatibility complex class II alleles Mamu-DRB1*1003 and -DRB1*0306 are enriched in a cohort of simian immunodeficiency virus-infected rhesus macaque elite controllers. J Virol 82:859–870. doi:.10.1128/JVI.01816-07 [PMC free article] [PubMed] [Cross Ref]
44. Harris M, Burns CM, Becker EA, Braasch AT, Gostick E, Johnson RC, Broman KW, Price DA, Friedrich TC, O'Connor SL 2013. Acute-phase CD8 T cell responses that select for escape variants are needed to control live attenuated simian immunodeficiency virus. J Virol 87:9353–9364. doi:.10.1128/JVI.00909-13 [PMC free article] [PubMed] [Cross Ref]
45. O'Connor SL, Lhost JJ, Becker EA, Detmer AM, Johnson RC, Macnair CE, Wiseman RW, Karl JA, Greene JM, Burwitz BJ, Bimber BN, Lank SM, Tuscher JJ, Mee ET, Rose NJ, Desrosiers RC, Hughes AL, Friedrich TC, Carrington M, O'Connor DH 2010. MHC heterozygote advantage in simian immunodeficiency virus-infected Mauritian cynomolgus macaques. Sci Transl Med 2:22ra18. doi:.10.1126/scitranslmed.3000524 [PMC free article] [PubMed] [Cross Ref]

Articles from Infection and Immunity are provided here courtesy of American Society for Microbiology (ASM)