![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Previous Article | Next Article 
Infection and Immunity, December 2006, p. 6865-6876, Vol. 74, No. 12
0019-9567/06/$08.00+0 doi:10.1128/IAI.00561-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
University of Arkansas for Medical Sciences, Central Arkansas Veterans Healthcare System, Little Rock, Arkansas,1 Case Western Reserve University School of Medicine,2 Louis Stokes Cleveland Veterans' Affairs Medical Center,3 University Hospitals of Cleveland, Cleveland, Ohio4
Received 5 April 2006/ Returned for modification 12 June 2006/ Accepted 6 September 2006
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recently, use of models based on intracellular infection of human phagocytes has demonstrated that patterns of intracellular growth of reference strains of M. tuberculosis correlate with the observed virulence of these isolates in experimental animals (22, 31). Studies have also suggested that intracellular growth of M. tuberculosis correlates with clinical evidence of virulence as suggested by the capacity of clinical isolates to cause outbreaks of disease (25, 32). The development or lack of development of outbreaks may involve factors other than strain virulence, however, such as the infectiousness of an index case, the number of individuals exposed and the intensity of their exposures, and the health status and resulting susceptibility of exposed individuals.
Unfortunately, when disease is not transmitted by an apparently infectious index case, information regarding these potentially confounding issues is often limited. Lack of such data hinders the selection of appropriate low-virulence "control" isolates for use in these assays. However, efforts to identify the biological features unique to unusually virulent isolates of M. tuberculosis require comparison to control strains for which lower virulence is well established on epidemiological grounds. In addition, without the identification of such isolates, it is difficult to verify the ability of in vitro models to distinguish between strains of varying virulence.
In the current study, we used a case-control design to identify M. tuberculosis isolates for which there was evidence of transmission of the organism and isolates from comparable tuberculosis cases in which infection was not transmitted. Isolates were obtained from an ongoing study of tuberculosis patients (i.e., "index cases") in Kampala, Uganda, and their household contacts (5, 16). We identified households (HH) in which transmission of tuberculosis occurred in three scenarios: (i) presence of coprevalent (CP) disease, defined as development of active disease in a HH member within 60 days of identification of the index case, (ii) development of incident (IC) disease by HH members 6 months or more after identification of the index case, and (iii) development of infection (IF) without active disease in HH contacts. We then searched the HH contact database to identify control isolates of M. tuberculosis from other matched Kampala index cases in which transmission of the organism was not observed. Matching was based both on clinical characteristics of the index patient and upon HH size, living conditions, and health status of other HH members.
In vitro studies of virulence were performed using two human assay systems that were developed independently in our two laboratories. One of these utilizes the human macrophage-like THP-1 cell line. In this model, THP-1 cells are preactivated with phorbol myristate acetate to induce adherence and with gamma interferon, which enhances their capacity for phagocytosis. Infection of THP-1 cells with M. tuberculosis has been performed using a 50:1 multiplicity of infection to induce easily detectable cytokine responses. In our hands, this level of infection has not led to significant loss of cell viability over a 1-week assay on intracellular growth (26). However, both intracellular growth and patterns of cytokine induction by clinical isolates of M. tuberculosis have shown correlation with epidemiologic evidence of strain virulence (25). The other virulence model utilizes primary peripheral blood monocytes (MN). In these cells, M. tuberculosis infection has been performed with a 1:1 multiplicity of infection to approximate clinical exposure to M. tuberculosis and to specifically eliminate differences in cell viability following infection with virulent and avirulent laboratory strains of the organism. Adequate phagocytosis is achieved by performing infections in medium using fresh autologous donor serum rather than by preactivation of the cells. Intracellular growth of laboratory strains H37Ra, Mycobacterium bovis BCG, and H37Rv within primary MN displayed patterns of growth that correlate with their virulence in animal models of infection (22). In addition, infection of MN in a blinded fashion using a panel of clinical isolates of M. tuberculosis indicated that similar rapid growth was displayed by isolates that shared the spoligotype of the rapidly transmitted Beijing isolate family (11). In the current study, the organisms identified using the epidemiologic criteria described above were provided in a blinded fashion to investigators at both study sites. In both THP-1 and primary MN models, intracellular growth of M. tuberculosis correlated with epidemiologic evidence of strain virulence.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
5-mm induration in response to 5 TU of M. tuberculosis purified protein derivative (PPD) in HH members who displayed no symptoms or radiographic evidence of active tuberculosis. For each HH in which evidence of transmission of M. tuberculosis was observed, the database was searched to identify appropriate control HH in which no transmission of the organism was observed. Epidemiologic information regarding a total of 302 tuberculosis patients and their respective HH monitored during the 3-year period of the project was encoded and, using SAS, compared to that of each of the 9 HH in which transmission was observed. Matching was based on index case characteristics relevant to transmission of disease (i.e., age category, acid-fast bacillus [AFB] smear grade, presence or absence of cavitary disease, duration of cough, and HIV status), HH characteristics relating to intensity of exposure (i.e., number of persons per room), and HH member characteristics relevant to tuberculosis risk (age category, body mass index [BMI], history of BCG vaccination, and HIV status). The search identified nontransmission control HH with the closest overall matches to the transmission HH with regard to all of the epidemiologic measures noted. In instances when multiple nontransmission HH were identified as equally well matched to the transmission HH, a single HH was randomly selected as the source for the control isolate.
For each of the three transmission scenarios, isolates from three HH were selected for in vitro study based of the availability of a well-matched control HH in which there was no evidence of transmission of the organism. In vitro studies therefore involved 9 pairs of M. tuberculosis isolates in which each pair consisted of one transmitted isolate (T) and one nontransmitted control isolate (C).
Each transmission M. tuberculosis isolate obtained from HH members with coprevalent and incident disease was subjected to restriction fragment length polymorphism (RFLP) analysis using IS6110 and spoligotyping as previously described (8, 27). In each HH, genotyping confirmed intra-HH transmission by demonstrating that isolates obtained from index and contact cases within the HH were identical. In addition, multiple isolates from each index case were subjected to IS6110 RFLP analysis to confirm that the isolate selected for study was representative of the organism responsible for that patient's disease. For each index case, the RFLP patterns of each isolate were identical. Molecular analysis also indicated that that transmitted and nontransmitted isolates represented distinct M. tuberculosis strains (Fig. 1).
|
Model of infection with the THP1 human macrophage-like cell line. Human macrophage-like THP-1 cells were grown in RPMI 1640 with 10% heat-inactivated fetal calf serum, 25 mM HEPES, and 2 mM glutamine in 5% CO2 at 37°C. THP-1 cells were activated into adherent macrophages by the addition of 100 nM phorbol myristate acetate and recombinant 100 U/ml human gamma interferon for 1 to 3 days at 37°C until used in experiments, as previously described (26). Cells were aliquoted into 96-well plates (2 x 105 cells/well), washed twice in binding medium (138 mM NaCl, 8.1 mM Na2HPO4, 1.5 mM KH2PO4, 2.7 mM KCl, 0.6 mM CaCl2, 1 mM MgCl2, and 5.5 mM D-glucose), and acclimated for 10 min before addition of each of the 18 strains in a 50:1 multiplicity of infection. Tubercle bacilli and macrophages were gently rocked for 1 h at 37°C, 5% CO2, followed by 2 h of stationary incubation. After infection, monolayers were washed twice with phosphate-buffered saline to remove extracellular bacteria. Three hours after infection and 7 days subsequently, cell supernatants were removed and cell pellets lysed with 0.1% sodium dodecyl sulfate, as previously described (26). Serial 10-fold dilutions of cell lysates were then prepared in 7H9 medium. For each dilution, triplicate 10-µl aliquots of lysate were plated onto 7H10 agar. Intracellular growth was expressed as the growth ratio (day 7 CFU divided by day 0 CFU). Despite the high initial infection ratio, the viability of THP-1 cells as measured by trypan blue exclusion was not adversely impacted, as 71 to 94% of cells remained viable at end of the 7-day assay for each of the 18 strains studied.
Human peripheral blood monocyte model of infection. In these studies performed at Case Western Reserve University, peripheral blood was obtained from healthy Cleveland-area volunteers. Potential subjects were excluded from participation if they were receiving antibiotics or any type of immunosuppressive medications. Mononuclear cell fractions were obtained by density sedimentation over Ficoll-Hypaque. Peripheral blood monocytes (MN) were then purified via plastic adherence, as previously described (21). MN were washed, counted, and resuspended in Iscove's modified Dulbecco's medium (IMDM) plus 10% fresh autologous serum (AS) and 1% penicillin G at a density of 1 x 106 per ml. Aliquots of 100 µl were plated in triplicate into 96-well flat-bottom plates and incubated at 37°C overnight to allow adherence. The next morning, MN were incubated with the various M. tuberculosis isolates. Incubations were performed using a 1:1 multiplicity of infection in IMDM with 30% AS and 1% penicillin G. After incubation at 37°C for 1 h, nonphagocytosed organisms were rinsed from the plates and medium was replaced with IMDM with 10% AS and 1% penicillin G. Both immediately after infection and 7 days subsequently, cell supernatants were removed and cell pellets lysed with 0.1% sodium dodecyl sulfate, as previously described (21). Serial 10-fold dilutions of cell lysates were then prepared using 7H9 medium. For each dilution, triplicate 10-µl aliquots of lysate were plated onto 7H10 agar (with OADC enrichment). When visible colonies were observed, CFU were counted under a microscope. Growth was again expressed as the ratio of CFU at day 7 to initial CFU at day 0.
Cytokine determinations.
Culture supernatants from infected THP-1 macrophages were harvested after association for 3 h (day 0) and on days 1, 3, 5, and 7, frozen at –70°C, and then assayed with commercial enzyme-linked immunosorbent assay (ELISA) kits according to the manufacturer's instructions as previously described (25). Tumor necrosis factor alpha (TNF-
) and interleukin-10 (IL-10) were not detectable in supernatants of cells cultured in the absence of M. tuberculosis.
Cytokine production by infected peripheral blood MN was assessed by plating 1 x 105 MN per well into flat-bottom 96-well plates. Following overnight incubation, MN of each of the donors were simultaneously infected in the manner described above with each of the three coprevalent transmitted isolates and the corresponding three isolates from matched HH in which coprevalent disease was not observed. Culture supernatants were collected 24 and 48 h after infection for assessment of TNF- production, and after 1, 3, 5, and 7 days of culture for measurement of IL-10 induction. Supernatants from MN of 10 subjects were collected and frozen at –70°C until the time of their use in commercially available ELISA kits as previously described (11).
For both infection models, the TNF- bioactivity of culture supernatants was also assessed using the L929 cytotoxicity assay. For these experiments, 100 µl of a suspension of 4 x 105 L929 murine fibroblasts (ATCC, Rockville, MD) was seeded in RPMI 1640 with 10% fetal calf serum onto 96-well flat-bottom plates. After removal of culture medium, 100 µl of each supernatant, run in duplicate, was added to the cells with 100 µl of medium containing actinomycin D (5 µg/ml) for 24 h. The extent of cell death was determined by crystal violet staining and absorbance at 600 nm. Culture medium alone and serial dilutions of recombinant TNF- were used as negative controls and assay standards, respectively.
Statistical analysis. Statistical analysis of differences in intracellular growth was based on comparison of log-transformed growth ratios to satisfy the normal distribution assumption for the parametric statistical methods used in the analysis. Comparisons therefore utilized the unit log growth ratio, defined as the log of the ratio of CFU count at day 7 over CFU count at day 0 (equivalent to the difference between the log CFU counts at days 7 and 0). The general linear and mixed effects statistical models with categorical (or fixed effect) strain were used to analyze sets of paired strains for the THP-1 and primary human MN experiments, respectively (10). Preliminary exploratory analyses were conducted separately on the nine "transmitted" strain/nontransmitted "control" strain pairs.
(i) Analysis of THP-1 experiments. Three replicate independent experiments were performed to measure the growth ratio for each of the 18 strains in this study for a total of 54 experiments. The independent group t test with three replications per strain was utilized to calculate P values for each comparison between the mean log growth ratio of transmitted and nontransmitted control isolates for each of the nine strain pairs. Analysis of the overall difference in intracellular growth between the three paired isolates of each clinical scenario was performed using a general linear model that properly adjusted for the matching of strain pairs.
Statistical analysis of THP-1 cell cytokine production was performed using paired t test or analysis of variance. P values of <0.05 were considered significant.
(ii) Analysis of primary human monocyte experiments. MN of each healthy Cleveland-area donor were infected with each of the 18 studied strains in a single separate experiment. Performance of these studies using MN from 12 donors resulted in a total of 216 growth ratio determinations. Comparisons of the log growth ratio of each of the nine transmitted and nontransmitted strain pairs were performed using paired t tests to adjust for correlation among log growth ratios observed in MN taken from a single donor. Assessment of the overall differences in growth between the three transmitted and control isolates of each clinical scenario utilized a mixed effects model that adjusted for pairing as well as for infection of MN from multiple donors with each clinical isolate. For these analyses, the donor was treated as a random effect and pairing of case and matched strains was treated as a fixed effect. Covariance between repeated measurements of growth ratios within donors was adequately described using the compound symmetry model.
Statistical analysis of MN cytokine production was performed using paired t tests.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
3 months). HH characteristics were then evaluated to identify the best-matched control HH in which coprevalent disease, incident disease, or infection in the absence of active disease, respectively, were not observed. Isolates are identified according to clinical scenario (CP, IC, and IF) and by whether the HH represented was one in which there was evidence of disease transmission (T), or a matched control HH (C) in which evidence of transmission was not observed.
|
|
|
3 months, whereas the index case of nontransmission control HH CP2-C had cough for <3 months. In coprevalent HH CP1-T, there were two cases of coprevalent disease, both in underweight 1-year-old children (defined as those with a body mass index of <18.5) who had not received BCG. In coprevalent HH CP2-T, coprevalent disease was observed in an underweight 3-year-old child who had not received BCG. An incident case was subsequently observed in an adult HH member of normal weight. In transmission HH CP3-T, coprevalent disease was found in an underweight 1-year-old child who had not received BCG. An incident case was subsequently observed in an underweight 12-year-old child as well. In all three coprevalent case pairs, the transmission and nontransmission control HH were well matched in terms of numbers of HH contacts, HH crowding, and distribution of highly susceptible individuals as defined above. IC transmission HH were defined as those in which an HIV-negative HH contact developed active tuberculosis more than 6 months after the baseline evaluation of the index case. As summarized in Table 2, the index cases of incident disease transmission and nontransmission control HH were comparable with regard to age group and clinical characteristics. In transmission HH IC1-T and IC3-T, incident disease was observed in otherwise healthy adults, whereas in HH IC2-T, a malnourished adult developed incident disease. The HH were also well matched for numbers of very young children and for the presence underweight individuals. None of the IC transmission or control HH members were HIV positive. Of note, prior to the development of incident disease, the transmission HH of IC pair 2 was selected as a matched HH for infection pair 2 (Table 3).
IF transmission HH were identified as those in which all contacts displayed positive skin test responses to PPD (defined as an induration of 5 mm or more) at the time of identification of the index case, whereas in nontransmission control HH, all contacts had negative skin tests using this same cutoff. In transmission HH IF1-T, PPD responses of HH members ranged in size from 5- to 15-mm induration. In transmission HH IF2-T, the range of skin test responses was 6 to 17 mm of induration, whereas in transmission HH IF3-T, the range was 15 to 22 mm of induration. As detailed in Table 3, infection pair index cases were exactly matched the by presence or absence of cavitary disease, AFB smear grade, coughing category, and age group. Likewise, the transmission and nontransmission control HH for each pair were equivalent in terms of HH crowding. The HH were similar in distribution of highly susceptible HH members (very young children, HIV-infected individuals, and malnourished individuals) as well. For this group of subjects, however, basing selection criteria for control HH on PPD reactivity may have underestimated the potential virulence of the strains isolated from each of these HH. Specifically, in nontransmission control HH IF1-C, one HIV-positive family member had developed a 3-mm induration in response to PPD skin testing. Although this did not reach the 5-mm cutoff used in the definition of infection, a reaction of this size in an HIV-infected individual could indicate infection with M. tuberculosis. Control HH IF2-C had no skin-test-positive subjects but did have a member who subsequently developed an incident case of active tuberculosis, as noted above.
Intracellular growth of M. tuberculosis isolates in THP-1 cells and in primary human monocytes. All paired clinical isolates of M. tuberculosis were provided to the investigators without labeling as to which clinical scenario they represented or whether the isolate was from a HH with evidence of transmission or from a nontransmission control HH. The isolates were unblinded for statistical analysis only after all CFU data had been reported.
The intracellular growth of the 9 pairs of M. tuberculosis isolates was expressed as the growth ratio (CFU at day 7 divided by CFU at day 0) for both THP-1 and primary human MN models of infection. These results are shown in Fig. 2.
|
For the three IC disease strain pairs, the THP-1 model again indicated significant differences in growth between pairs of transmission and nontransmission control HH isolates. The growth of the transmission HH isolate was greater than isolate from the control HH in each pair (P = 0.0031, P = 0.0002, and P < 0.0001 for IC pairs 1, 2, and 3, respectively), and the overall P value for the group of paired IC isolates was statistically significant (P < 0.0001). In the primary MN model, the growth of transmission HH isolates IC1-T and IC2-T was more rapid than that of the isolates from corresponding nontransmission isolates, but these differences were not statistically significant for this number of subjects (P = 0.22 and P = 0.13, respectively). The growth of the transmitted IC3-T isolate was slightly less rapid than that of control isolate IC3-M, but again this difference was not statistically significant (P = 0.58). The combined analysis of intracellular growth of isolates from all three pairs of transmitted and nontransmitted control HH in the MN model, again, did not indicate statistically significant differences (P = 0.16).
For the IF pairs, transmitted isolate IF2-T displayed greater growth than the corresponding nontransmitted control isolate in the THP-1 model (P = 0.021), whereas matched isolates IF1-C and IF3-C had greater intracellular growth than their respective transmitted isolates in this system (P = 0.035 and P = 0.0037, respectively). In the primary human MN model, isolates from each of the three IF pairs displayed relatively similar growth and the differences observed did not approach statistical significance (with P values for comparisons between IF pairs 1, 2, and 3 of 0.68, 0.97, and 0.60, respectively). When growth was analyzed for the entire set of paired IF isolates, the overall differences between isolates from transmitted and control HH were not significant in either the THP-1 or primary MN models (P = 0.17 and P = 0.99, respectively).
In the combined analyses, therefore, both THP-1 and primary human MN models demonstrated significantly greater growth of isolates from HH with CP disease compared to nontransmitted control HH. Both models suggested greater growth of isolates from IC transmission HH than from control HH, but these differences were significant in the THP-1 model only, whereas neither model demonstrated significant differences in the intracellular growth of IF transmission isolates and their nontransmitted controls.
Cytokine induction by paired isolates of M. tuberculosis.
In the THP-1 model, we have previously observed a strong correlation between the intracellular growth of M. tuberculosis and patterns of cytokine induction. Specifically, isolates that grow rapidly in this system are poor inducers of TNF- and induce IL-10 responses more rapidly than do more slowly growing isolates (25). Because the THP-1 and MN models both demonstrated differences in intracellular growth between M. tuberculosis isolates from HH with coprevalent disease and their nontransmitted control HH, these six paired isolates were used to evaluate the correlation between TNF- and IL-10 responses and intracellular growth.
In the THP-1 model system, peak TNF- levels and bioactivity for each isolate were observed 48 h following infection. Results are shown in Table 4. As seen in each pair of M. tuberculosis isolates, the organism from the transmission HH induced significantly less TNF- than did the corresponding control HH isolate. The activity of TNF-, as measured by the L929 bioassay, corresponded with the observed protein levels in these supernatants, as shown.
|
were observed 24 h after infection (Fig. 3A). Comparisons of the results of infection of MN from each of 10 donors with paired organisms indicate that coprevalent transmission isolates CP1-T and CP3-T both tended to induce less TNF- than the isolates from control HH in which coprevalent disease was not observed. Due to donor-to-donor variability, these results are not statistically significant, as the P values were 0.067 and 0.113 by paired t test, respectively. Coprevalent transmission isolate CP2-T tended to induce more TNF- than its matched isolate, although this finding was again not statistically significant (P = 0.060). TNF- data from 48 h after infection showed similar patterns, but again, the differences observed were not statistically significant (data not shown). TNF- bioactivity within the same culture supernatants was assessed as well (Fig. 3B). Again, wide subject-to-subject variability was observed. As was seen with supernatants of infected THP-1 cells, the relative levels of TNF- activity induced by the paired CP strains paralleled the ELISA results, but for the MN model, bioactivity was substantially lower than the corresponding ELISA results for each isolate. The levels of TNF- bioactivity from MN infected with transmitted strain CP3-T were significantly lower than those of MN infected with the corresponding nontransmitted control strain CP3-C (P = 0.026 by paired t test), whereas differences in TNF- bioactivity between the other coprevalent pairs remained nonsignificant.
|
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the THP-1 model, differences in intracellular growth were observed between each pair of clinical isolates, whereas this was not the case with primary cells. This finding may in part reflect the superior reproducibility of the THP-1 model of infection. Reproducibility allows for detection of statistically significant differences in intracellular growth using only a limited number of replicate experiments. As such, this method provides a more practical means for screening a large number of isolates to identify those with unusual capacity for rapid intracellular growth. When the three paired isolates in each scenario were analyzed as a group, the coprevalent and incident transmission HH isolates demonstrated significantly greater growth within THP-1 cells than those of their nontransmitted control HH, whereas infection isolates did not.
In the primary MN model, growth of the coprevalent transmission isolates CP2-T and CP3-T demonstrated significantly greater growth than isolates from nontransmitted control HH. When analyzed as a group, the intracellular growth of the three coprevalent transmission isolates was also significantly greater than that of the control HH isolates. Transmitted isolates from incident disease HH displayed more rapid growth than those of their control HH as well, although these differences were not statistically significant, whereas growth of the infection transmission HH isolates and their nontransmitted controls were fairly similar.
Both infection models, therefore, were capable of detecting differences in intracellular growth of clinical isolates of M. tuberculosis that corresponded to detailed epidemiological information regarding these isolates. These differences were significant for both models when comparing M. tuberculosis isolates from HH with coprevalent disease with isolates from control HH in which only the index case demonstrated active disease. These findings may be consistent with the relative virulence of the isolates as reflected in these scenarios in that the finding of coprevalent tuberculosis cases presumably reflects the capacity of the organism to cause rapid progression from exposure and initial infection to active disease. The observation of incident disease suggests somewhat delayed progression of this process, which could explain the less robust differences observed in the intracellular growth of these paired isolates in the primary MN model.
The finding of infection did not display a clear correlation with intracellular growth. The lack of overall differences in the intracellular growth of the infection strain pairs may reflect the limitations of the criteria (i.e., PPD skin test responses) used for selecting this set of isolate pairs. For example, one skin-test-negative member of an IF control HH subsequently developed active tuberculosis that was confirmed by RFLP to be due to the index organism, indicating that this HH member was incorrectly classified as "uninfected." Factors such as malnutrition, coexisting parasitic disease, and HIV infection could account for false-negative skin test results in this setting, as could the performance of skin testing relatively soon after infection, prior to the development of specific cell-mediated immunity (7). In addition, the utility of skin testing as a indicator of virulence has been questioned by previous studies suggesting that increased incidence of skin test conversion in patient contacts may instead reflect strain immunogenicity (14). More generally, the use of PPD testing as a measure of infection is problematic in populations such as ours in which BCG vaccination is common. In future studies, the use of antigen-specific ELISAs such as QuantiFERON-TB Gold is likely to provide a more accurate measure of recent infection with M. tuberculosis (2, 15).
The two infection models also yielded different results in terms of the relationship between intracellular growth and cytokine profiles (of TNF- and IL-10), as assessed in studies of the CP transmission and nontransmitted control isolates. In the THP-1 model, transmitted isolates that displayed more rapid intracellular growth consistently induced less TNF- than did slower-growing control isolates. Likewise, in the THP-1 model, rapidly growing strains induced earlier production of IL-10 than control isolates. These results coincide with those previously reported for clinical strains associated with outbreaks, i.e., rapid production of IL-10 and suppression of TNF- in the THP-1 model are highly correlated with the rapid growth phenotype (25). In the MN model, transmitted isolate CP3-T induced less bioactive TNF- than its nontransmitted control isolate CP3-C. None of the other comparisons of MN cytokine production were statistically significant, as is frequently the case with primary cells given the wide range of donor-to-donor variability of such responses. However, coprevalent transmission isolates CP1-T also appeared to tend toward induction of less TNF- than did matched nontransmitted control isolates, whereas the opposite appeared to be true of transmitted isolate CP2-T and the control isolate. In all cases, TNF- bioactivity paralleled ELISA results for TNF-, although it did appear that inhibitors of TNF- activity were induced by M. tuberculosis in the MN system but not in the THP-1 model. With regard to IL-10, coprevalent transmission isolate CP2-T appeared to have a tendency toward more-rapid IL-10 induction than its matched isolate, whereas CP1-T and CP3-T induced very little cytokine. Together, these results could be interpreted to suggest that either lower TNF- production or more-rapid induction of IL-10 is required to facilitate rapid intracellular growth of more virulent isolates of M. tuberculosis. Given widely varying results previously reported for such assessments (12, 13, 22, 25, 32), however, another possibility may be that differences in induction of TNF- and IL-10 are not the primary determinants of strain growth. This interpretation would be consistent with other studies implicating possible roles of the differential abilities of M. tuberculosis strains to evade phagocyte apoptosis (9) and to interfere with phagolysosome fusion (18, 30) in the organisms' capacity for intracellular growth.
In addition to these considerations, it must be noted that the methodologies of the two models differ in significant ways that clearly could impact the outcomes of intracellular infection. These include the inherent differences between a differentiated tumor cell line and primary cells obtained from peripheral blood, as well as the requirement for preactivation of THP-1 cells, and the use of both different types of serum and different multiplicities of infection in the two systems. With regard to these differences, it should be emphasized that our goal was to compare the outcomes of infection within these two assays as they have been used in previously published studies, rather than to modify the assays to make the protocols more similar. The observation that dissimilar in vitro models both demonstrate significant differences in growth of isolates obtained from the most "aggressive" clinical scenario, that of coprevalent disease, and their nontransmitted control isolates further supports the concept that capacity for intracellular growth plays a central role in the virulence of M. tuberculosis.
The goal of studies such as this one is identification of paired clinical isolates of M. tuberculosis that have biological differences relevant to their respective virulence. Determination of whether one assay is more "sensitive" or "specific" with regard to this goal, therefore, ultimately requires further study of the organisms themselves. In any case, our findings support the concept that the ability of M. tuberculosis to evade defenses of the phagocyte and thus display increased capacity for intracellular growth is a primary characteristic of rapidly transmitted isolates. Our observation that the individuals who developed active disease most rapidly following HH exposure were very young children may thus suggest that such phagocyte defenses are not fully developed in this age group.
In summary, the use of case-control methods in our study has confirmed the ability of two infection models to demonstrate differences in intracellular growth of M. tuberculosis isolates that correlate with epidemiological evidence of virulence. Paired isolates identified in this manner should be of interest for further studies aimed at clarifying virulence mechanisms of M. tuberculosis. This may be particularly true of isolates obtained from Africa due to the finding that isolates obtained from this continent show relatively less genetic diversity than those from elsewhere (4). Accordingly, differences that can be identified among these strains may be more likely to be related to biologic differences relevant to their virulence.
|
ACKNOWLEDGMENTS |
|---|
This study was supported by the Tuberculosis Research Unit (NO1-AI-95383).
|
FOOTNOTES |
|---|
Published ahead of print on 18 September 2006.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
| 1. | Alland, D., G. E. Kalkut, A. R. Moss, R. A. McAdam, J. A. Hahn, W. Bosworth, E. Drucker, and B. R. Bloom. 1994. Transmission of tuberculosis in New York City-an analysis by DNA fingerprinting and conventional epidemiologic methods N. Engl. J. Med. 330:1710-1716. |
| 2. | Brock, I., K. Weldingh, T. Lillebaek, F. Follmann, and P. Andersen. 2004. Comparison of tuberculin skin test and new specific blood test in tuberculosis contacts. Am. J. Respir. Crit. Care Med. 170:65-69. |
| 3. | Collins, F., and M. Smith. 1969. A comparative study of the virulence of Mycobacterium tuberculosis measured in mice and guinea pigs. Am. Rev. Respir. Dis. 100:631-639.[Medline] |
| 4. | Filliol, I., A. Motiwala, M. Cavatore, W. Qi, M. Hazbon, M. Bobadilla del Valle, J. Fyfe, L. Garcia-Garcia, N. Rastogi, C. Sola, T. Zozio, M. Guerrero, C. Leon, J. Crabtree, S. Angiuoli, K. Eisenach, R. Durmaz, M. Joloba, A. Rendon, J. Sifuentes-Osornio, A. Ponce de Leon, M. Cave, R. Fleischmann, T. Whittam, and D. Alland. 2006. Global phylogeny of Mycobacterium tuberculosis based on single nucleotide polymorphism (SNP) analysis: insights into tuberculosis evolution, phylogenetic accuracy of other DNA fingerprinting systems, and recommendations for a minimal standard SNP set. J. Bacteriol. 188:759-772. |
| 5. | Guwatudde, D., M. Nakakeeto, E. C. Jones-Lopez, A. Maganda, A. Chiunda, R. D. Mugerwa, J. J. Ellner, G. Bukenya, and C. C. Whalen. 2003. Tuberculosis in household contacts of infectious cases in Kampala, Uganda. Am. J. Epidemiol. 158:887-898. |
| 6. | Guwatudde, D., S. Zalwango, M. R. Kamya, S. M. Debanne, M. I. Diasz, A. Okwera, R. D. Mugerwa, C. King, and C. C. Whalen. 2003. Burden of tuberculosis in Kampala, Uganda. Bull. W. H. O. 81:799-805.[Medline] |
| 7. | Hanson, C. A., and L. B. Reichman. 1989. Tuberculosis skin testing and preventative therapy. Semin. Respir. Infect. 4:182-188.[Medline] |
| 8. | Kamerbeek, J., L. Schouls, A. Kolk, M. van Agterveld, D. van Soolingen, S. Kuijper, A. Bunschoten, H. Molhuizen, R. Shaw, M. Goyal, and J. van Embden. 1997. Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J. Clin. Microbiol. 35:907-914.[Abstract] |
| 9. | Keane, J., H. Remhold, and H. Kronfeld. 2000. Virulent Mycobacterium tuberculosis strains evade apoptosis of infected alveolar macrophages. J. Immunol. 164:2016-2020. |
| 10. | Laird, N., and J. Ware. 1982. Random effects models for longitudinal data. Biometrics 38:963-974.[CrossRef][Medline] |
| 11. | Li, Q., C. C. Whalen, J. M. Albert, R. Larkin, L. Zukowski, M. D. Cave, and R. F. Silver. 2002. Differences in rate and variability of intracellular growth of a panel of Mycobacterium tuberculosis clinical isolates within a human monocyte model. Infect. Immun. 70:6489-6493. |
| 12. | Lopez, B., D. Aguilar, H. Orozco, M. Burger, C. Espitia, V. Ritacco, L. Barrera, K. Kremer, R. Hernandez-Pando, K. Huygen, and D. Van Soolingen. 2003. A marked difference in pathogenesis and immune response induced by different Mycobacterium tuberculosis genotypes. Clin. Exp. Immunol. 133:30-37.[CrossRef][Medline] |
| 13. | Manca, C., M. B. Reed, S. Freeman, B. Mathema, B. Kreiswirth, C. E. Barry III, and G. Kaplan. 2004. Differential monocyte activation underlies strain-specific Mycobacterium tuberculosis pathogenesis. Infect. Immun. 72:5511-5514. |
| 14. | Manca, C., L. Tsenova, C. E. Barry III, A. Bergtold, S. Freeman, P. A. J. Haslett, J. M. Musser, V. H. Freedman, and G. Kaplan. 1999. Mycobacterium tuberculosis CDC1551 induces a more vigorous host response in vivo and in vitro, but is not more virulent than other clinical isolates. J. Immunol. 162:6740-6746. |
| 15. | Mazurek, G., J. Jereb, P. LoBue, M. Iademarco, B. Metchock, and A. Vernon. 2005. Guidelines for using the QuantiFERON-TB gold test for detection of Mycobacterium tuberculosis infection, United States. Morb. Mortal. Wkly. Rep. 54(RR15):49-55.[Medline] |
| 16. | Mudido, P., D. Guwatudde, M. Nakakeeto, G. B. Bukenya, D. Nsamba, J. L. Johnson, R. D. Mugerwa, J. J. Ellner, and C. C. Whalen. 1999. The effect of bacille Calmette-Guerin vaccination at birth on tuberculin skin test reactivity in Ugandan children. Int. J. Tuberc. Lung Dis. 3:891-895.[Medline] |
| 17. | North, R., and A. Izzo. 1993. Mycobacterial virulence: virulent strains of Mycobacteria tuberculosis have faster in vivo doubling times and are better equipped to resist growth-inhibiting functions of macrophages in the presence and absence of specific immunity. J. Exp. Med. 177:1723-1733. |
| 18. | Pethe, K., D. Swenson, S. Alonso, J. Anderson, C. Wang, and D. Russell. 2004. Isolation of Mycobacterium tuberculosis mutants defective in the arrest of phagosome maturation. Proc. Natl. Acad. Sci. USA 101:13642-13647. |
| 19. | Pierce, C., R. Dubos, and M. Schaefer. 1953. Multiplication and survival of tubercle bacilli in the organs of mice. J. Exp. Med. 97:189-206.[Abstract] |
| 20. | Public Health Service. 1985. Public health mycobacteriology: a guide for the level III laboratory. Public Health Service, U. S. Department of Health and Human Services, Washington, D.C. |
| 21. | Silver, R., Q. Li, W. Boom, and J. Ellner. 1998. Lymphocyte-dependent inhibition of growth of virulent Mycobacterium tuberculosis H37Rv within human monocytes: requirement for CD4+ T cells in purified protein derivative-positive, but not in purified protein deriviative-negative subjects. J. Immunol. 160:2408-2417. |
| 22. | Silver, R. F., Q. Li, and J. J. Ellner. 1998. Expression of virulence of Mycobacterium tuberculosis within human monocytes: virulence correlates with intracellular growth and induction of tumor necrosis factor alpha but not with evasion of lymphocyte-dependent monocyte effector functions. Infect. Immun. 66:1190-1199. |
| 23. | Small, P. M., P. C. Hopewell, S. P. Singh, A. Paz, J. Parsonnet, D. C. Ruston, G. F. Schecter, C. L. Daley, and G. K. Schoolnik. 1994. The epidemiology of tuberculosis in San Francisco-a population-based study using conventional and molecular methods. N. Engl. J. Med. 330:1703-1709. |
| 24. | Steenken, J., and L. U. Gardner. 1946. History of H37 strain of tubercle bacillus. Am. Rev. Tuberc. 54:62-66. |
| 25. | Theus, S., M. Cave, and K. Eisenach. 2005. Intracellular macrophage growth rates and cytokines of Mycobacterium tuberculosis strains with different transmission dynamics. J. Infect. Dis. 191:453-460.[CrossRef][Medline] |
| 26. | Theus, S. A., M. D. Cave, and K. D. Eisenach. 2004. Activated THP-1 cells: an attractive model for the assessment of intracellular growth rates of Mycobacterium tuberculosis isolates. Infect. Immun. 72:1169-1173. |
| 27. | van Embden, J., M. Cave, J. Crawford, J. Dale, K. Eisenach, B. Gicquel, P. Mermans, C. Martin, R. McAdam, T. Shinnick, and P. Small. 1993. Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: recommendations for a standardized methodology. J. Clin. Microbiol. 31:406-409. |
| 28. | van Rie, A., R. Warren, M. Richardson, T. C. Victor, R. P. Gie, D. A. Enarson, N. Beyers, and P. D. van Helden. 1999. Exogenous reinfection as a cause of recurrent tuberculosis after curative treatment. N. Engl. J. Med. 341:1174-1179. |
| 29. | van Soolingen, D., L. Qian, P. E. de Haas, J. T. Douglas, H. Traore, F. Portaels, H. Z. Qing, D. Enkhsaikan, P. Nymadawa, and J. D. van Embden. 1995. Predominance of a single genotype of Mycobacterium tuberculosis in countries of east Asia. J. Clin. Microbiol. 33:3234-3238.[Abstract] |
| 30. | Vergne, I., J. Chua, H. Lee, M. Lucas, J. Belisle, and V. Deretic. 2005. Mechanism of phagolysosome biogenesis block by viable Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 102:4033-4038. |
| 31. | Zhang, M., J. Gong, Y. Lin, and P. Barnes. 1998. Growth of virulent and avirulent Mycobacterium tuberculosis strains in human macrophages. Infect. Immun. 66:794-799. |
| 32. | Zhang, M., J. Gong, Z. Yang, B. Samten, D. M. Cave, and P. F. Barnes. 1999. Enhanced capacity of a widespread strain of Mycobacterium tuberculosis to grow in human macrophages. J. Infect. Dis. 179:1213-1217.[CrossRef][Medline] |
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| J. Bacteriol. | J. Virol. |
|---|
