Immunodiagnosis Of Active Tuberculosis Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Tuberculosis disease is a major health problem in the world today. The causative agent of the disease Mycobacterium tuberculosis infects individuals by inhalation of M. tb aerosols produced by a person with the disease, mainly through coughing, or sneezing. When the bacteria enter the lungs, the person's innate immunity is activated. They bind to phagocytic receptors and are taken up by the macrophages, dendritic cells and monocytes that have been recruited from the blood stream (Bhatt et al., 2007). In a small number of individuals the bacteria is cleared from the system, but in the majority of individuals the activation of the adaptive immunity is required to keep bacterial growth in check. If the bacteria are not successfully cleared, this results in a persistent infection state which is known as latent TB infection (LTBI) (Bhatt et al., 2007). Individuals that are latently infected with M. tb are not contagious; do not show any symptoms of disease and most of them will not develop active TB disease in their lifetime even if untreated. A small number of individuals are never latently infected as they immediately progress to active disease (Bhatt et al., 2005).

Approximately one third of the world population is reported to be latently infected with M. tb (WHO, 2009). Tuberculosis disease is highly prevalent in the third world where the rate of latent tuberculosis infection is also high. Individuals who are latently infected have a greater risk of developing post-primary active tuberculosis disease especially those who have compromised immune systems such as co-infection with human immunodeficiency virus (HIV), malnutrition and treatment with immuno-suppressive drugs (Goletti et al., 2006). Approximately 10% of people who are latently infected with M. tb will develop active tuberculosis disease during their lifetimes. At that time, they will show symptoms of disease and be able to spread the disease (Goletti et al., 2006).

1.2 Immune response to M. tuberculosis infection

The immune response towards M. tb infection is predominantly cell mediated immunity. M. tb is phagocytosed by macrophages, these macrophages then present antigens of the intracellular bacteria on their MHC class II complex. T helper (Th) 1 cells that bind to the MHC class II complex that presents the antigens on the outside of the macrophages then secrete cytokines in response to activate the macrophages and B-cells (Kaufmann et al., 2008). The T cells produce interferon-γ, which activates the macrophages so they can control M. tb replication. They also secrete other host markers such as interleukin (IL) 17, which protect against TB by participating in the early steps of granuloma formation, tumour necrosis factor alpha (TNF-α), which functions to activate macrophages, granulocyte-macrophage colony-stimulating factor (GM-CSF) stimulates granulocyte and macrophage maturation and Interleukin 6 (IL-6) functions as a pro-inflammatory mediator (Kaufmann et al., 2008). B-cells and thus antibody production is stimulated by Th2 cells that secrete IL-4 and IL-5 (Kaufmann et al., 2008).

Tuberculosis is mainly a pulmonary disease, but in some patients, especially those with compromised immune systems, more cases of extra-pulmonary TB are being detected (Golden et al., 2005). The infection sites in extra-pulmonary tuberculosis (EPTB) includes all organs except the lungs, but the most common sites include the lymph nodes, pleura and osteoarticular areas (Golden et al., 2005). Because of the wide variety of infection sites found in EPTB there is also a wide variety of clinical symptoms, this together with the rarity of the disease makes it more challenging to diagnose compared to pulmonary TB (PTB) (Nishimura et al., 2008).

1.3 Diagnostic methods

Since the 1930's the tuberculin skin test (TST) has been the main test used for the diagnosis of tuberculosis infection (Janssens et al., 2007). Although the test is used in some areas to help in the diagnosis of active tuberculosis disease (A-TB) it is mostly recommended for latent tuberculosis infection (LTBI), as it cannot differentiate between LTBI and A-TB (Vincenti et al., 2003). Other limitations of the test include the fact that it cannot differentiate between infection with M. tb and other non-tuberculous mycobacteria, and may give false positive results in individuals that have been Bacille Calmette-Guérin (BCG) vaccinated (Bellete et al., 2002). The sensitivity of the TST test is also reduced in infected individuals that are at higher risk of progressing to active disease such as immuno-compromised individuals, children and elderly persons (Bakir et al., 2008).

Interferon-Gamma Release assays (IGRA's) were recently developed to diagnose LTBI. They work on the basis that the Th 1 cells of persons infected with M. tb will recognize M. tb specific antigens and secrete IFN-γ in response to these antigens. The levels of IFN-γ are then measured and compared to a positive and negative control. The advantages of these assays over the TST is that the number of false positives are far less because of the M. tb specific antigens that are used, the results are more rapid and the interpretation of results of IGRA's are not as subjective as the results of the TST.

In low prevalence countries where they treat individuals with LTBI and A-TB, the need to distinguish between the two states of infection is not as important. In contrast, countries with a high prevalence of LTBI, and which often have limited resources, focus their treatment programs on individuals with A-TB so as to limit the spread of disease, thus a need to be able to rapidly and effectively differentiate between individuals with A-TB from those with LTBI has arisen (Vincenti et al., 2003).

The diagnosis of active tuberculosis disease up to date has relied on clinical symptoms, radiological, microbiological and pathological information gathered for each patient by a medical professional (Chen et al., 2009). The most widely used method for the diagnosis of active TB is the examination of a Ziehl Neelson stained smear for acid fast bacilli. Although this method is rapid, not technically very demanding and relatively inexpensive it is not very sensitive for M. tb and a high concentration (5000 to 10 000) of cells per millilitre of blood is needed to confirm the diagnosis (Steingart et al., 2006). Mycobacterial culture is considered to be the reference standard for diagnosis of TB disease. Culture is more sensitive than smear microscopy, but culture on the Lowenstein Jenson medium is slow and can take anything from 15 to 40 days (Vincenti et al., 2003), which means the onset of the patients treatment is delayed and this also prolongs the time that the patients can potentially infect other individuals. The time to culture positivity is however reduced if liquid culture systems are used, but these are more expensive, requires more technical skills to perform and culture facilities are rarely available in most poor/high burden TB settings, another shortcoming of this method is the fact that in only about 50% of cases M. tb can be cultured (Kang et al., 2007) with even less in individuals who are HIV+ (Vincenti et al., 2007).

All the shortcomings of the above mentioned methods for the diagnosis of tuberculosis has recently led to research being focused on the development of more rapid and effective methods to diagnose patients with active TB disease. These methods include the so-called Interferon-Gamma Release Assays (IGRA's) and also the serodiagnosis of TB by testing for M. tb specific antibodies in the patients, usually through ELISA (Enzyme-linked Immunosorbent Assay) based methods. The focus of this review is to give an update on recent developments in the immunodiagnostic methods that have been attempted for the diagnosis of A-TB.

2. IGRA's

2.1 Principles

There are currently two commercially available Interferon-γ Release Assays that are widely used for the diagnosis of M. tb infection. These include the T-SPOT.TB test (Oxford Immunotec, Oxfordshire, Abington, UK) and the Quantiferon tests (Cellestis, Carnegie, Australia). The T SPOT.TB test is an ELISPOT based method and the Quantiferon Gold tests are ELISA based technique. Both tests are based on the principle that in individuals who are infected with M. tb, there are pre-activated T cells circulating in their blood. When the pre-activated T cells from these individuals re-encounter M. tb specific antigens, they respond rapidly by the production of protective cytokines and other biomarkers. IGRA's measure IFN-γ that has been produced after stimulation of these cells in vitro. The T-SPOT.TB assay involves the enumeration of T-cells that secrete IFN-γ upon stimulation with M. tb specific antigens while the Quantiferon assays measures the total IFN-γ response of peripheral blood mononuclear cells (PBMC's) to these M. tb specific antigens.

In the T SPOT.TB assay, PBMC's are isolated from whole-blood. These cells are then stimulated with early secretory antigenic target 6 (ESAT-6) and culture filtrate protein 10 (CFP-10) individually on the plate and a negative (saline) and positive (phytohaemagglutinin (PHA)) control is done for each sample (Kang et al., 2007). After overnight incubation (16 to 20 hours), cells that have previously been exposed to M. tb will recognize the TB specific antigens will secrete IFN-γ in response. The IFN-γ secreting cells are counted in the TB specific antigen stimulated wells and then compared to those obtained in the unstimulated and positive control wells to ascertain the result (positive, negative or indeterminate result).

For the Quantiferon Gold In-Tube assay 1ml of whole blood is drawn into three tubes a nil, a TB specific antigen and mitogen tube. The nil tube is the negative (unstimulated) control tube and contains only saline. The TB specific antigen tube contains a combination of ESAT-6, CFP-10 and TB7.7 (Rv2654), and the mitogen tube, which contains PHA, serves as a positive control. These tubes are then incubated overnight (16 to 24hours) after which the supernatants are harvested. The amount of IFN-γ that has been produced in response to stimulation by the contents of each tube is then determined by ELISA (Kang et al., 2007; Ferrara et al., 2006).

In some studies in-house varieties of these tests have been done that use different antigens (e.g. HBHA (Tuuminen et al., 2007) or different peptides and thus different epitopes of the antigens used (Goletti et al., 2006). The results of these studies mostly correlate with the findings of studies using the commercial tests regarding sensitivity and specificity.

2.2 Use of IGRA's in the diagnosis of A-TB

IGRA's were originally developed for the diagnosis of LTBI, and because they are so specific for M. tb infection recent research has focused on trying to modify them for the diagnosis of A-TB to improve their usefulness in a clinical setting.

Goletti et al. evaluated the accuracy of the Quantiferon Gold and T SPOT.TB compared to ELISPOT assay using RD1 selected epitopes of ESAT-6 and CFP-10 as antigens. The study population consisted of adults, who were HIV- with confirmed TB disease in Rome, Italy. They found the Quantiferon to have a specificity of 59% and sensitivity of 83% while the T SPOT.TB performed better with a specificity and sensitivity of 59% and 91% respectively (Goletti et al., 2006). The RD1 selected epitope assay used in this study performed even better with a sensitivity and specificity of 70% and 91% (Goletti et al., 2006). In another study Goletti et al. found the RD1 selected epitope assay to be much more specific than the T SPOT.TB and the Quantiferon Gold In-Tube (specificities of 70.6%, 44.3% and 61.9% respectively) (Goletti et al., 2008).

Cattamanchi et al. examined the performance of T SPOT.TB in diagnosing PTB in patients who are HIV+. They found the that the assay could correctly identify 60% of patients with A-TB, but this is only in the patients with an adequate number of PBMC's and who showed an good mitogen response (Cattamanchi et al., 2010). They nonetheless found a sensitivity and specificity of 73% and 54% (Cattamanchi et al., 2010).

In a study done in Japan they assessed the ability of the Quantiferon Gold assay to differentiate between pulmonary TB (PTB) and extra-pulmonary TB (EPTB). They found a sensitivity of 74.5% for detecting PTB and 80.0% for EPTB with an overall specificity of 91.2% (Nishimura et al., 2008). The sensitivities in this study for detecting A-TB was low but the specificity for detecting PTB and EPTB did not differ significantly (Nishimura et al., 2008).

A number of other studies have been done to determine the specificity and sensitivity of the IGRA's in high and low incidence settings as well in different population groups but they have mostly found results that do not differ significantly. They found sensitivities varying between 83% and 100% (Cattamanchi et al., 2010; Goletti et al., 2006; Rangaka et al., 2007; Vincenti et al., 2007) and specificities between 65% and 95% (Goletti et al., 2006; Vincenti et al., 2007).

When the IGRA's are done on samples from children the results often vary from studies done on adults (Bakir et al., 2008). In a study done by Bamford et al. found that IGRA's are generally less sensitive in children than in adults with Quantiferon Gold In-tube having a sensitivity of 78% and the T SPOT.TB a sensitivity of only 66% (Bamford et al., 2009), this is contradictory to many studies that find the T SPOT to be more sensitive in studies done with adults (Janssens et al., 2007; Winqvist et al., 2009; Dominguez et al., 2008).

Chegou et al. recently found that the best single markers to differentiate between individuals with A-TB and those without A-TB disease were EGF (epidermal growth factor) and sCD40L (soluble CD40 ligand), with other models using combinations of three markers being even more sensitive and specific in differentiating A-TB from non-ATB (Chegou et al., 2009). They found that a combination of EGFNil, EGFAg-Nil and MIPAg-Nil correctly predicted the highest percentage of individuals with ATB, by identifying 96% of ATB cases (Chegou et al., 2009).

Bronchoalveolar lavage mononuclear cells (BALMC's) obtained from subjects with suspected PTB were used to perform ELISPOT assay, using ESAT-6 and CFP-10 as antigens, in a study done by Jafari et al. They found T cells specific to M. tb to be more concentrated at the site of infection than in the blood (Jafari et al., 2008). The authors suggest that if the concentration of BALMC's are compared to those of PBMC's it may be possible to accurately differentiate between A-TB and LTBI.

3. Serodiagnosis of tuberculosis

If an individual is infected with M. tb there are antibodies and B-cells specific for M. tb antigens present in the body no matter where the infection is located, this makes serological tests a promising alternative to the present methods available to diagnose A-TB especially EPTB that can sometimes be hard to diagnose as such a high percentage of such cases cannot be cultured to confirm the diagnosis. These methods are also advantageous because they are rapid to perform and it is easier the get the samples needed to perform these tests than it is to get adequate sputum samples from children for culturing

It is hypothesised that the antigen composition of M. tuberculosis changes when latent infection progresses to active disease (Davidow et al., 2005). Theoretically this makes it possible to design serological tests that can identify individuals that are latently infected or that have active TB disease. If the correct combination of antigens are chosen, in other words antigens that are immunogenic and mainly expressed in M. tuberculosis bacilli that are in the exponential growth phase (Bukhary, 2007; Okuda et al., 2004), it would be possible to diagnose individuals that have A-TB using serological assays.

Serological tests have been used for the diagnosis of tuberculosis since 1898 when crude cell preparations from M. tb were used as antigens (Singh et al., 2007) Since then the field has made developed much further and today ELISA's and rapid immunochromatographic assays are used (Singh et al., 2007).

3.1 Principles

Most of the existing serodiagnostic tests for tuberculosis are in the form of ELISAs (Tiwari et al., 2005; Bukhary, 2010; Anderson et al., 2008; Okuda et al., 2004). In these assays, a microtitre plate is coated with the antigen or capture antibody of choice and serum, plasma or other bodily fluid samples are added to the wells. If an individual is infected with the M. tb the antibodies against the different antigens of M. tb will be present in the serum. The antibodies used in the ELISA can be native or recombinant antibodies. Upon addition of the sample from an infected individual, these antibodies (in the patient's samples) will bind to the antigens that coat the wells. After blocking of the plate, these antibodies will stick to the plate when it is washed, while the rest of the sample is washed away. Anti-human antibodies that are specific for a certain class of antibody, eg. IgG, are then added to each well and allowed to bind to the antibodies that bound to the antigens (Tiwari et al., 2005). The anti-human antibodies (secondary antibodies) are conjugated to a substrate that gives off a colour when it reacts to a certain enzyme (Tiwari et al., 2005). Unbound secondary antibodies are washed away, enzyme is added and the intensity of the colour reaction that is formed is then used as an indicator of the amount of antibodies the individual had against the antigen.

As briefly mentioned in the previous section, the rapid serological tests are based on the principle of immune chromatography. The test sample (serum, plasma or whole-blood may be used) is placed on a membrane, as the sample flows through the membrane by chromatographic principles, the coloured recombinant TB antigen-colloidal gold conjugate complexes with the TB specific antigens that are present in the sample (Tripathi et al., 2004). The bound complex then moves further on the membrane until the TB specific antigens immobilises the complex and a colour formation reaction takes place (Tripathi et al., 2004). The formation of a coloured band indicates a positive result, while the absence of one means a negative result, any unbound complexes moves further on the membrane to the control region where it is bound by anti-rabbit antibodies (Tripathi et al., 2004). This binding then causes a pink colour formation and serves as a positive control (Tripathi et al., 2004).

3.2 Ability of serological tests to differentiate between A-TB and LTBI

Most serological tests are designed to detect individuals with active TB disease and not those with LTBI (Bukhary, 2007; Okuda et al., 2004). Therefore, most serodiagnostic studies are not designed and named specifically as studies to discriminate between active TB and LTBI. The abilities of these tests to discriminate between A-TB and LTBI is therefore judged by their sensitivity (for the TB cases studied) and specificity (in the non active-TB cases evaluated).

Singh et al. (2009) found that the serum of individuals that were PPD (tuberculin skin test)-positive and thus "latently infected" did not have antibodies against four peptides of the proline-threonine repetitive protein (PTRP) that they used as antigens in a serological assay. In contrast individuals with A-TB had a strong immunogenic response to the peptides used (Singh et al., 2009). They found that the four immunodominant peptides used in the assay could identify more than 80% of smear positive TB cases and more than 50% of smear-negative TB cases that were used in this study, while none of the PPD positive controls tested positive (Singh et al., 2009).

Okuda et al. (2003) used three commercially available serological tests namely (tuberculous glycolipids (TBGL), lipoarabinomannan (LAM) and antigen (A) 60 antibody detection assays and combined the results to make a diagnosis. They found a combined specificity of 83.8% and a sensitivity of 87.5% when the results of these three tests were combined (Okadu et al., 2003).They found a high number of false positives that may be due to the fact that the different antigens they used are not only expressed in M. tb during exponential growth and because their study group was located in an area with a high prevalence of latent M. tb infection (Okuda et al., 2004).

Most of the commercially available serological assays are based on these traditional antigens (LAM, A60, TBGL, Heat shock protein 65). The results that were obtained in earlier serological assays (especially the rapid tests) often showed very high specificity for A-TB (above 95%), but yielded poor sensitivities (less than 33%). This is typical of findings that were obtained with rapid tests such as the ICT tuberculosis test. A meta-analyis conducted by Steingart and co-workers, showed that accuracies of these serological tests was very poor, and that they could not be recommended for routine use (Steingart et al., 2008). The most recently developed assays however, seem to be showing promise. Rajan et al. evaluated the 65kD heat shock protein in serum samples for the diagnosis of A-TB by an ELISA assay with monoclonal antibodies specific for this protein. They found a sensitivity and specificity of 82% and 89% respectively (Rajan et al.,2007). This looks to be a promising assay to be utilised for the diagnosis of A-TB.

Raqib and co-workers (Raqib et al., 2009), evaluated the accuracy of recently developed B cell-based assay in paediatric samples. The patients comprised individuals with confirmed pulmonary, glandular, intestinal, spinal TB and TB meningitis, with the majority having pulmonary TB. The sensitivity of the test for A-TB was 92% and the specificity was 87.5% (Raqib et al., 2009). Khan et al. (2008) isolated PBMC's and stimulated them with Rv1168c or PPD and then measured the IFN-γ and IL-5 response after 4 days. They found that this antigen elicited a strong antibody response in individuals with A-TB (both PTB and EPTB) and it was also found to be a potent T-cell antigen which increases the amount of IFN-γ that is secreted by PBMC's (Khan et al., 2008).

Individuals that are infected with HIV are at high risk of developing A-TB, studies using IGRA's are not very promising in diagnosing these individuals are these assays rely on the IFN-γ secretion of T-cells and the HIV virus affects T-cells and lowers the number of cells found in vivo. This is where serodiagnosis of A-TB can play an important role as the production of antibodies are a function of B-cells which are theoretically not affected by HIV infection.

However Anderson et al. showed that of the three commercially available assays, namely InBios Active TbDetect IgG ELISA, the IBL M. tuberculosis IgG ELISA and the Anda Biologicals TB ELISA they used only two (InBios Active TbDetect IgG ELISA and IBL M. tuberculosis IgG ELISA) could reliably detect antibodies against M. tb antigens in serum samples in individuals co-infected with HIV and M. tb.

An immune response to the four peptides of the PTRP protein used in the study by Singh et al. in contrast elicited an immune response that could by detected in more than 80% of their study population that where co-infected with HIV and M. tb.

Because of the diversity of antibodies found in different individuals not all subjects will have antibodies against a specific antigen. For this reason there should be tested for antibodies against more than one antigen specific to M. tuberculosis to improve the sensitivity of the assay (Okuda et al., 2004; Tiwari et al., 2005; Raqib et al., 2009).

M. tb also also presents the host with different antigen profiles during different phases of infection which makes it difficult to develop a test that can detect all individuals with TB disease (Kaufmann et al., 2008; Raqib et al., 2009).

4. Discussion

The development of methods to accurately diagnose A-TB rapidly and non-invasively would be a major advance in the field of TB diagnostics. As of yet the only commercial tests that are available can only reliably detect LTBI, although they achieve this with great sensitivity and specificity, and cannot with great confidence differentiate between LTBI and A-TB. The need has thus arisen to either modify these commercial tests so they can effectively detect A-TB disease or to develop new tests altogether.

IGRA's and serological tests at this time point seem to be the best methods to be utilised in the fight against the worldwide TB pandemic.

In studies done with IGRA's the results show that these assays are very specific for M. tb, but their sensitivity is not good enough for them to be utilised in a clinical setting for the diagnosis of A-TB as currently are and further research is needed to improve the sensitivity (Goletti et al., 2006; Goletti et al., 2005; Janssens et al., 2007; Winqvist et al., 2009). These findings are in line with those found in children although it seems that where the T SPOT.TB is usually more specific en sensitive in adult study populations (Dominguez et al., 2008; Bamford et al., 2009), the Quantiferon assays give better sensitivity and specificity in paediatric studies (Bamford et al., 2009; Bakir et al., 2008). They also reported a general lower sensitivity of IGRA's in children, this could be attributed to the fact that children show less robust immune responses to M. tb specific antigens in vitro just as they do when they become infected with M. tb in vivo (Bamford et al., 2009). The reason for thus less effective defence against M. tb is that young children have less TB specific T cells (T cells secreting IFN-γ in response to stimulation to TB specific antigens) and a more generalised T cell suppression during A-TB (Bamford et al., 2009).

Because IGRA's are T cells based tests and the CD4 T cells are the cells that mainly secrete IFN-γ the use of these assays with severe immuno-compromised patients seem to be limited, especially individuals that are HIV+ (Cattamanchi et al., 2010).

The IGRA's especially the Quantiferon Gold assay seem to be just as sensitive and specific for diagnosing EPTB than it is diagnosing PTB (Nishimura et al., 2008), this is a promising result as it would make the diagnosis of EPTB much easier and less invasive. It could also lead to more cases of TB being detected as EPTB is often misdiagnosed because of the wide variety of symptoms associated with the many different types if EPTB (Nishimura et al., 2008). While Jafari et al. conducted their study on PTB their results could be applied to EPTB as well. They found that there are more M. tb specific T cells at the site of infection than there is circulating in the blood of an infected individual (Jafari et al., 2008). If samples can thus be obtained from the site of infection in cases with EPTB a more accurate diagnosis can be made and other infections can be ruled out with greater confidence.

Chegou et al. found that IFN-γ may not be the best biomarker of M. tb infection to test for when trying to diagnose A-TB. When a combination of biomarkers are used, A-TB cases are diagnosed more accurately than when only using IFN-γ (Chegou et al., 2009).

Overall the IGRA's are promising assays because of they specificity for M. tb infection, but they cannot as of yet be used in clinical settings because of their inability to reliably differentiate between LTBI and A-TB. One reason for this low sensitivity is the lack of a gold standard to differentiate between LTBI and A-TB. Further research needs to be done, to improve the sensitivity of these assays fot A-TB, especially in high incidence settings.

Serological testing has come a long way since its discovery in the 1800's. This diagnostic method is an attractive method to diagnose A-TB, because it is non-invasive, rapid and not a high level of technical skills and expensive laboratory equipment is needed to perform them.

Unlike the IGRA's these assays are mostly not developed for the diagnosis of LTBI, but for A-TB. The major problem encountered when using these assays is the fact that a combination of many antigens must be used because of the diverse antibody profile an individual can have against M. tb (Raqib et al., 2009; Khan et al., 2008; Anderson et al., 2008). This profile is also different for each individual which makes it even more challenging to develop an assay that is sensitive enough in all or most individuals.

Further research is thus needed to try and find antigens or combinations of antigens that are only expressed during A-TB and which can be recognized by the antibody profiles of most individuals that have active TB disease.