I/St and A/Sn mice are polar extremes in terms of several parameters defining sensitivity to Mycobacterium tuberculosis. TNF-α, mainly produced by activated macrophages, can mediate both physiological and pathophysiological processes. Adequate TNF-α levels are essential for a forceful protective response to M. tuberculosis. We have functionally characterized a nonsynonymous substitution, Arg8His, in the highly conserved cytoplasmic domain of the pro-TNF-α leader peptide from extremely M. tuberculosis-sensitive I/St mice. This was compared to the common pro-TNF-α variant found in A/Sn mice. Using cDNA constructs, both variants were constitutively expressed in HEK293A cells. A significantly higher secretion level of Arg8His TNF-α was shown using flow cytometry and ELISA analysis (P=0.0063), while intracellular levels were similar for both protein variants. An even TNF-α distribution throughout the cells was seen using confocal microscopy. This suggests that the Arg8His substitution affects pro-TNF-α processing. The I/St mouse may serve as a model to further explore the function of the well-conserved cytoplasmic region of TNF-α. However, other identified substitutions in the I/St promoter, introns and 3′UTR of Tnf-α, as well as the cellular environment in vivo may affect the balance between soluble and intracellular Arg8His TNF-α before and during M. tuberculosis infection.
I/St inbred mice display an extremely rapid disease progression after intravenous challenge with virulent human isolate Mycobacterium tuberculosis H37Rv, whereas the A/Sn inbred mouse strain is that most resistant to disease progression among over 30 inbred strains tested.1 Compared to A/Sn mice, the I/St mice display a two times faster body weight loss, a 20- to 100-fold higher mycobacterial multiplication rate in spleens and lungs, and a more severe lung histopathology and less than half of the survival time.2, 3 These strains constitute a tool to study genetic factors controlling sensitivity to tuberculosis since they clearly differ in a number of parameters defining disease progression. Linkage analysis in crosses between I/St and A/Sn mice suggested a quantitative trait locus on chromosome 17, which implies an underlying gene that controls the variation in severity of disease among the offspring.2, 4 This locus contains the highly polymorphic histocompatibility-2 locus, H-2, where the cytokine gene Tnf-α is located.
TNF-α is a pleiotropic proinflammatory cytokine,5 which seems to be able to affect almost all cell types.6 It can mediate both physiological and pathophysiological processes, and was reported to be involved in cachexia,7 regulation of programmed cell death8 and play a significant role in septic shock.9 TNF-α secretion by macrophages, dendritic cells and T cells is induced during infection with M. tuberculosis.10 It is clear that the TNF-α response to M. tuberculosis infection is important for the infection progression in mice,11, 12, 13 as well as in humans, strongly supported by the fact that administration of TNF-α blockers enhances the risk for M. tuberculosis infection reactivation.14 Variations in the human Tnf-α gene promoter have been extensively studied for involvement in infectious disease, mostly the −238, −276 and −308 SNPs, in assays on TNF-α production and association studies. Results have been contradictory, and it is not yet clear if the actual SNPs under study are functional.15, 16 However, fairly large association studies in this context have shown an association of these SNPs with susceptibility to cerebral malaria or severe malarial anemia.15 Since (i) excessive inflammatory responses are central in the progression of tuberculosis, (ii) TNF-α is known to have an important role in the control of inflammation, and (iii) the rapid disease progression in M. tuberculosis infected I/St mice can in part be accounted for by the QTL containing the Tnf-α gene, TNF-α is a reasonable candidate to contribute to the difference in disease sensitivity between I/St and A/Sn. We hypothesized that there would be different TNF-α alleles in the I/St strain than in the A/Sn strain resulting in dissimilar TNF-α activity. Accordingly, we found several inter-strain variations in the Tnf-α gene, including a variation giving rise to an amino-acid substitution in the conservative cytoplasmic region of the pro-TNF-α leader sequence.
The TNF-α precursor is a 26 kDa type II transmembrane protein (pro-TNF-α), present as a homotrimer,17 which is transferred to a 17 kDa soluble form by proteolytic cleavage of the leader sequence.18, 19, 20 The transmembrane and the soluble forms were shown to be able to trigger biological activities distinct from one another.18, 21, 22, 23, 24, 25 Proteolytic cleavage of TNF-α is therefore believed to be important in the regulation of TNF-α's diverse functions. A difference in this processing, or in protein localization, between the two mouse strains may account for part of the large difference in response to M. tuberculosis challenge seen between the I/St and A/Sn mice. We hypothesized that the detected amino-acid substitution in the I/St strain would affect secretion of the soluble TNF-α. To test that, we examined secretion, intracellular level, and localization of the new I/St and the common A/Sn TNF-α variants by constitutive expression in HEK293A cells.
We identified a substitution in the Tnf-α gene encoding a novel leader peptide sequence in extremely M. tuberculosis sensitive I/St mice and studied its functionality. The secreted and intracellular levels of the novel protein were compared with the levels of the A/Sn variant, by the analysis of fluorescent TNF-α-fusion proteins expressed in HEK293A cells. The I/St and the A/Sn TNF-α variants were identical but for this substitution.
Identification of a new Arg8His polymorphism
The Tnf-α gene was sequenced in M. tuberculosis sensitive I/St (H2j) and insensitive A/Sn (H2a) inbred mouse strains. A novel nonsynonymous single base pair substitution was found in I/St but not A/Sn at +23 (with respect to the translational start site, NCBI Accession number Y00467) in exon 1 (Arg8His ((A/Sn) G>A (I/St))). The amino-acid substitution was situated in the cytoplasmic region of the leader sequence in pro-TNF-α. To determine if this substitution is present within other mouse strains, we searched the Genbank Mouse (mouseblast.informatics.jax.org), finding Tnf-α sequences for 12 additional strains: BALB/c (H2d), C57BL/6 (H2b), C3H/HeJ (H2k), P/J (H2p), SJL (H2s), SWR/J (H2q), NOD (H2g7), CTS (H2ct), BFM/2Msf, A/J (H2a), 129 (H2b), B10.M (H2f). All these strains had the arginine amino acid at position eight, and the A/Sn variant was therefore considered as the common variant among mice. To determine if the substitution is present within strains with similar haplotype as I/St, we sequenced strains available from Jackson Laboratory with the H2j haplotype as well as C57BL/10 (H2b) (B10). The H2-congenic strain B10.WB-H2jH2-T18b/SnJ (B10.WB) but neither C3.JK-H2jH2-T18b/SnJ (C3.JK) nor B10 contained the Arg8His substitution. B10.WB mice were more sensitive to M. tuberculosis H37Rv infection than B10 mice as revealed by the mean postinfection survival time. Introduction of H2j but not H2k or H2d haplotypes on the B10 background made the mice more infection sensitive (Table 1).
The amino-acid sequence of the exon 1 encoded peptide is highly conserved among mammals. When comparing the exon 1 sequence between mouse (NCBI Accession Number Y00467) with the other mammals available at the NCBI website, the following homology values (%) were found: rat (91.9), white-footed mouse (90.3), llama (88.7), cat (88.7), baboon (88.7), human (87.1), guinea pig (87.1), chimpanzee (87.1), horse (87.1), red deer (86.0), dog (85.5), woodchuck (85.5), baboon (85.5), rhesus macaque (85.5), pig (85.5), bovine (83.9), zebu (83.9), water buffalo (83.4), atlantic bottle-nosed dolphin (82.3), beluga whale (82.3), goat (82.3), rabbit (82.3), crab eating macaque (82.3), olive baboon (82.3), sheep (82.3), common squirrel monkey (78.7), tammar wallaby (71.0) and brush-tailed possum (67.7). At position eight, where the new amino acid in the I/St mice is situated, the amino acid is arginine in all mammals mentioned above, including humans, except in the crab eating macaque and common squirrel monkey where it is glutamine. This arginine residue is at position eight in all mammals but red deer, where it is at position three.
To our knowledge, the Arg8His substitution has not previously been reported, and no other SNP in the leader sequence, in either man or mouse, has previously been studied.
Arg8His substitution increased the secretion of TNF-α
The A/Sn and I/St construct variants were identical but for the Arg8His substitution. Cells with the Arg8His construct, that is, the I/St TNF-α variant, had higher secretion levels than cells with the common construct, that is, the TNF-α variant in for example A/Sn. Secretion level was corrected for differences in cell number between samples, and was defined as the supernatant TNF-α concentration divided by the ratio: number of fluorescent TNF-α-ZsGreen producing cells in the well/an average cell number per well. This average cell number was based on all A/Sn and I/St samples and was 52 787 cells and 30 711 cells at 72 and 48 h post-transfection, respectively. Multiple regression and two-way ANOVA of outcome variable secretion, using protein variant and post-transfection time point as predictors, showed a significant dependence on protein variant (coef (mean±s.e.)=3.50±1.24, 95% CI: 1.01–5.98, 1 df, F=8.0, P=0.0063) (Table 2). Post-transfection time point did not contribute significantly to the model (coef (mean±s.e.)=−2.01±1.25, 95% CI: −4.50 to 0.48, 1 df, F=2.6, P=0.11). The secretion dependence on protein variant was supported by the results of two-tailed t-tests. There was a clear tendency at both time points for cells with the Arg8His construct to secrete more TNF-α (t=2.03, 34 df, P=0.051, and t=2.00, 27 df, P=0.057 at 72 and 48 h, respectively) (Table 2).
The intracellular TNF-α-ZsGreen level (ie mean channel of fluorescence) was similar for cells transfected with either of the two protein variants (F=0.06, P=0.82), also when analyzing each time point separately (t=0.62, P=0.54 and t=1.10, P=0.28, at 72 and 48 h post-transfection, respectively) (Figure 1). No interaction between protein variant and time point was found (P=0.36). In agreement, two-way ANOVA of secretion corrected for differences in intracellular TNF-α-ZsGreen level, by dividing the secretion variable with the ratio : mean channel of fluorescence from the TNF-α-ZsGreen producing cells/average mean fluorescence channel number, the dependence of secretion on protein variant still held (coef (mean±s.e.)=1.90±0.74, 95% CI: 0.41–3.39, 1 df, F=6.5, P=0.013). This average mean fluorescence channel number was based on all A/Sn and I/St samples and was 1807 and 1226 at 72 and 48 h post-transfection, respectively. However, studying secretion corrected for intracellular differences at each time point separately, the difference was weaker (t=1.94, P=0.060 and t=1.64, P=0.11, at 72 and 48 h post-transfection, respectively).
Thus, the Arg8His substitution was found to affect secretion of TNF-α but not intracellular level. Proportion of fluorescent cells was similar between the strains (P=0.75, Table 2), but slightly lower 72 h than 48 h post-transfection (P=0.035, two-way ANOVA and two-tailed t-test) indicating lower mitosis rate or decreased cell survival among transfected than nontransfected cells.
Localization did not differ between the Arg8His (I/St) and common (A/Sn) TNF-α variants
The confocal microscopy analysis of cells transfected with Arg8His (I/St) and common (A/Sn) TNF-α-ZsGreen vectors suggested that the mutation does not affect the intracellular localization of the protein. Both protein variants were distributed evenly throughout the cytoplasm, seen after 72 h (Figure 2) and 48 h (not shown).
Other single base pair substitutions were found in the I/St Tnf-α gene
Additional DNA sequence differences in the I/St Tnf-α gene compared to the A/Sn variant included the following single base pair substitutions: one in the promoter (−82 [A>C], with respect to the transcriptional start site which is situated 156 bases upstream the CDS, NCBI Accession Number Y00467), and one synonymous in exon 1 (Q59Q [C>T], two in intron 1 (+348 [G>A], +459 [C>A]), one in intron 3 (+1220 [G>A]), three in 3′UTR (+1820 [A>G], +2264 [C>T], +2418 [T>C]), with respect to the translational start site. The H2-congenic strains B10.WB-H2jH2-T18b/SnJ (B10.WB) but not C3.JK-H2jH2-T18b/SnJ (C3.JK) contained the I/St variants of the promoter and exon 1. None of these substitutions was previously reported in mouse. However, a synonymous SNP in mouse was reported at +24, and an I7T substitution was reported in BALB/c and C57BL/6. Neither of these polymorphisms was present in I/St, A/Sn or B10. These polymorphisms were not examined in this study.
We report a new functional variation in the Tnf-α leader sequence present in the extremely M. tuberculosis-sensitive I/St inbred mice, which encodes a substitution situated in the cytoplasmic domain of pro-TNF-α. The new Arg8His (I/St) TNF-α variant, and the common (A/Sn) TNF-α variant, identical but for the Arg8His substitution, were constitutively expressed as fluorescent fusion proteins in HEK293A cells, for analysis of TNF-α secretion and intracellular level. The Tnf-α promoter and 3′UTR were not included in the expression system. The Arg8His protein was secreted to a higher extent compared with the common variant (P=0.0063). This was shown as a higher supernatant Arg8His TNF-α level adjusted for number of fluorescent TNF-α-ZsGreen producing cells. The mean level of fluorescence from intracellular TNF-α-ZsGreen did not differ between Arg8His and the common variant. In agreement, the higher secretion from Arg8His was still present after dividing the secretion variable with this intracellular level parameter (P=0.013).
In line with our hypothesis that the Arg8His substitution in TNF-α contributes to the impaired response to M. tuberculosis in the I/St mice, H2-congenic B10.WB mice, with the same H2j haplotype as I/St mice, also displayed co-occurrence between a histidine residue at amino-acid position eight and M. tuberculosis sensitivity as compared to a corresponding arginine in resistant B10 mice. Moreover, an elevated I/St TNF-α secretion level, as seen in our in vitro expression system, could through an increased soluble TNF-α level explain the signs seen in the I/St mice after M. tuberculosis infection. However, the secretion level in vivo could be affected by other parameters, such as the additional variations found in the I/St Tnf-α gene. The TNF-α level from fresh lung macrophages has by Majorov et al26 been positively correlated with resistance studying I/St and A/Sn mice. On the other hand, the TNF-α levels, produced by lung cell suspensions consisting of T cells, macrophages and dendritic cells from M. tuberculosis H37Rv intratracheally infected I/St and A/Sn mice, did not differ between the mouse strains within the first 3 weeks postchallenge despite a dramatic upregulation of IL-12 and IFN-γ levels in A/Sn cells than in I/St cells. The interstrain differences in IL-12 production were largely independent of the presence of H37Rv sonicate and was not due to interstrain difference in number of inflammatory cells.27 IL-12 acts on T cells, macrophages and dendritic cells, for example, by inducing IFN-γ production.28 IL-12 in combination with mycobacteria has been found to enhance TNF-α production in macrophages, mediated through IFN-γ induction.29 The presence of a similar TNF-α secretion level from A/Sn and I/St cells, in a system with markedly higher stimulation of A/Sn compared to I/St cells, therefore, opens up the possibility for increased I/St TNF-α secretion from the inflammatory lung cell in a nonstimulated environment. The differences in TNF-α secretion between A/Sn and I/St lung macrophages reported by Majorov et al26 might also have been affected by a possible difference in IL-12 activation.
The I/St mice display a two times faster body weight and life loss, a 20- to 100-fold higher mycobacterial multiplication rate in spleens and lungs, a more severe lung histopathology, in comparison with the A/Sn mice.3 It has been proposed that the amount of TNF-α at the site of infection determines if the cytokine is protective or destructive.30 Ruuls et al25 suggested a more prominent role for the soluble form in inflammation, but it has also been reported that the membrane form by itself is sufficient to mediate inflammation.31 TNF-α induces cell death and cachexia.7, 8 Soluble TNF-α is implicated to have a dominant role in endotoxic shock and even death following bacterial infection.25, 32, 33, 34
Our localization analysis suggests that the substitution does not affect the intracellular localization of the protein, since an even cytoplasmic distribution was seen for both protein variants. This result is in agreement with the findings on the TNF-α secretion, since the higher secretion of the I/St variant implies that the protein successfully reached the membrane for cleavage and secretion. The positive effect on the secretion is more likely due to some effect on the TNF-α processing events.
The metalloproteinase TNF-α-converting enzyme (TACE; a disintegrin and metalloproteinase17; ADAM17) proteolytically cleaves pro-TNF-α, and findings indicate that TACE is responsible for most of the release of soluble TNF-α.19, 20, 35 It is not yet clear how the recognition and cleavage mechanism is performed by TACE. The amount of noncleaved TNF-α present in the cell membrane has been proposed to be under strict control, possibly by a feedback mechanism, with several studies showing no accumulation of transmembrane TNF-α when expressing a mutated noncleavable variant.22, 25, 36 Others report an accumulation of noncleavable TNF-α, but still less than expected if all of the nonprocessed TNF-α would reside in the membrane.37 HEK293 cells possess an endogenous TACE activity.38, 39 It was suggested that enzyme inhibitors and the amount of soluble TNF-α present can affect the TNF-α expression and secretion in this cell system.38
TACE exists in both a membrane-bound and a truncated soluble form, and two independent studies suggest that TACE must be membrane-anchored to cleave its substrates. They also report that it appears like both TACE and its substrate must be expressed in the same cell for processing to occur.40, 41 Reddy et al40 suggested that only the catalytic domain, which is located in the extracellular region, is essential for the cleavage process, at least in response to PMA stimulation. It is not known what function the cytoplasmic domain of TACE has, but an interaction with the substrate cytoplasmic region in some way could explain the increased secretion level for the I/St protein reported by us.
However, translocation and release of soluble TNF-α has previously been shown, in the human form, to proceed even after deletion of all cytoplasmic charged residues including Arg8 and approximately 70% of the cytoplasmic domain.42 Still, that this region is highly conserved indicates functionality. The Arg8His substitution may affect the processing selectivity. Zheng et al35 have demonstrated that the extracellular domain of TNF-α contributes to the selective cleavage by TACE, by preventing one or several other metalloproteinases from processing the cleavage site. The replacement of Arg8 by His might alter the conformation of the TNF-α trimer allowing other enzymes to gain access to the processing site. Pro-TNF-α is cleaved beyond a leucin residue, at position number 80 in the leader sequence, but two alternative cleavage sites have been reported.21, 22, 43 If the Arg8His substitution affects protein structure, these alternative processing sites might also become more accessible. Arginine is to 99% in its charged form whereas only 10% of histidine is charged at pH 7. There are additionally nine charged amino acids in the cytoplasmic region of the transmembrane pro-TNF-α.44
Some of the other sequence variations found between A/Sn and I/St may account for differences in TNF-α levels, and response to infection. However, these variations are not included in our expression system. A possible involvement of these variations in the expression and processing of TNF-α in the I/St mice remains to be explored. One mutation was found in the promoter (−82 [A>C]), and three in 3′UTR (+1820 [A>G], +2264 [C>T], +2418 [T>C]). The 3′UTR was shown to be involved in the regulation of stimulated TNF-α biosynthesis by enhancing mRNA translation efficiency following LPS stimulation, through ‘derepression’ of translationally repressive elements. An UA-rich element therein and its flanking segments were reported to be fundamental for this enhancement of TNF-α production.45, 46 The UA-rich region was also suggested to confer mRNA instability47 and to harbor an RNA-protein binding segment,48 as well as a completely mouse-human conserved site present in several different inflammatory mediators.49 The +2264 substitution is situated downstream this UA-rich region. Between the +2264 and +2418 substitution another RNA-protein binding site was reported.48 The 3′UTR has also been reported to cooperate with the TNF-α promoter in the regulation of gene expression.50 The −82 substitution was close to the TATA box, but was not within any transcription factor binding sites (Transcription Element Search System; Schug and Overton;51 www.cbil.upenn.edu/tess).
In summary, we found the Arg8His TNF-α variant, present in M. tuberculosis-sensitive I/St mice, to be secreted to a higher extent than the common TNF-α variant, present in, for example, M. tuberculosis insensitive A/Sn mice. The intracellular level of these variants was similar. The I/St mouse may serve as a model to further explore the function of the well-conserved cytoplasmic region of TNF-α. However, further studies are needed to determine if and how the other substitutions in I/St Tnf-α and the cellular environment in vivo affect the balance between soluble and intracellular Arg8His TNF-α before and during M. tuberculosis infection.
Materials and methods
Livers from mouse strains I/StSnEgYCit (I/St), A/SnYCit (A/Sn), C57BL/10 and H2-congenic strains B10.WB-H2j H2-T18b/SnJ (B10.WB) and C3.JK-H2j H2-T18b/SnJ (C3.JK) (Jackson Laboratory, Bar Harbor, ME, USA) were used for DNA amplification. The genomic sequence of Tnf-α (−714 to +2442), from the promoter to the polyadenylation signal, was determined from A/Sn and I/St using duplicate Tnf-α T-vector constructs. The Tnf-α promoter and CDS sequence was determined also in C57BL/10, B10.WB and C3.JK genomic DNA. Construct DNA and agarose gel electrophoresis-purified PCR products were subjected to sequencing with an ABI Prism™ 377-96 Collection machine using the BigDye™ Terminator Cycle Sequencing kit (Applied Biosystems, Foster City, CA, USA) and a polyacrylamide gel. Sequences were analyzed with ABI Prism, Sequencing Analysis software. All fragments and constructs were sequenced independently at least twice. Database Uniprot was applied for protein feature and amino-acid variation data searches (srs.ebi.ac.uk/srsbin/cgi-bin/wgetz?-page+srsq2+-noSession) and protein homology searches (www.ebi.ac.uk/fasta33). Nucleotide variation and homology searches were performed at the NCBI SNP website (www.ncbi.nlm.nih.gov/SNP) and NCBI BLAST website (www.ncbi.nlm.nih.gov/BLAST), respectively.
HEK293 cells can be used as a model system to study TNF-α processing. The precursor protein is cleaved at the correct proteolytic site, releasing a biologically active soluble form, which provide evidence for an endogenous, constitutive TNF-α converting enzyme activity.38, 39 It is suggested that enzyme inhibitors and the amount of soluble TNF-α present can affect the TNF-α expression and secretion in this cell system.26 The HEK293A cell line was grown in DMEM containing sodium pyruvate, 1000 mg/l glucose and pyridoxine (Invitrogen, Carlsbad, CA, USA), supplemented with 10% FBS (Sigma-Aldrich, St Louis, MO, USA), 1% L-glutamine (100 × ; Invitrogen) and 0.5% penicillin–streptomycin (10 000 U/ml penicillin G sodium and 10 000 μg/ml streptomycin sulfate in 0.85% saline; Invitrogen). Cells were grown as monolayers, in a 75 cm2 culture flask (Sarstedt, Nümbrecht, Germany) containing 30 ml cell medium, maintained in a humidified atmosphere of 5% CO2 at 37°C. Cells were passaged by trypzination after 2–3 days, upon reaching 100% confluency. For storage, cells were placed in 3 ml of 95% DMEM (10% FBS, 0.5% PEST, 1% L-glutamine) mixed with 5% DMSO, in −140°C. The cell culture was tested for mycoplasma contamination at the National Veterinary Institute (Uppsala, Sweden).
Cloning of the Tnf-α gene into fluorescence vectors
Frozen liver samples (−80°C), from I/St and A/Sn mice, were homogenized in TRIZOL reagent, and total RNA was isolated using phenol–chloroform extraction according to the protocol provided by the manufacturer (Gibco™ Invitrogen). The RNA samples were treated with DNase using the DNA-free™ Kit from Ambion (TX, USA) to remove potentially contaminating DNA. Total RNA was reverse transcribed using the Super Script II RT protocol (Invitrogen), and ds Tnf-α cDNA was produced using the Expand High Fidelity PCR system (Roche, Basel, Switzerland) with the following PCR primers: IndexTerm5′-AGATCTCGAGCCACATCTCCCTCCAGAA-3′ and IndexTerm5′-CCCGGGATCCAGAGCAATGACTCCAAAGTA-3′ (Proligo, Boulder, CO, USA). For constitutive expression of fluorescent fusion proteins, Tnf-α cDNA from position −26 to +705 was cloned N-terminal to ZsGreen in the green fluorescence eukaryotic expression vector pZsGreen1-N1 (BD Biosciences, Erembodegem, Belgium), and the plasmid DNA was purified using the S.N.A.P.™ midiprep kit (Invitrogen). Clones were sequenced as described above. Alignments with the control sequence (NCBI Accession Number Y00467) were performed using the GeneJockey II software (Biosoft, Cambridge, UK).
Before transfection, HEK293A cells were seeded in 24-well plates, or onto coverslips if the cells were transfected for confocal microscopy analysis. Cells were recovered from the T-flask by trypzination, and approximately 105 and 5 × 104 cells were added to each well for flow cytometry and confocal microscopy analysis, respectively. DMEM (10% FBS, 0.5% PEST, 1% L-glutamine) was added to a total volume of 1 ml. The plate was maintained in a humidified atmosphere of 5% CO2 at 37°C, for 24 h before transfection, which was conducted with the Lipofectamine™ Reagent, in conjunction with the Plus™ Reagent, using the protocol provided by the manufacturer (Invitrogen). Transfections were performed with the A/Sn and I/St expression vectors, and all experiments were conducted with the two construct variants in parallel. Cells transfected with the empty vector and a T-vector, which is nonfluorescent, served as positive and negative controls, respectively. The medium was exchanged approximately every 24 h until fixation.
Determination of time point of analysis after transfection
Time point for Tnf-α expression analysis was determined in initial experiments. The cell viability and the cell count, before and at different time points after transfection, were estimated using hemocytometer-based trypan blue dye (0.4% weight/volume in PBS; Sigma-Aldrich) exclusion. Two days post-transfection, cells had reached cell numbers higher than prior to transfection and the cell viability was 90% or above. Therefore, 48 and 72 h were chosen as time points for analysis.
After trypzination, suspended cells were fixed on ice with 3% PFA and 5% sucrose for 15 min, at time points 48 and 72 h post-transfection. After fixation, PBS was added, diluted samples were centrifuged for 3 min at 400 g, the supernatants were removed, whereafter cells were resuspended in 500 μl PBS. Single-cell suspensions were each mixed with a TruCOUNT™ tube containing a known number of fluorescent beads allowing estimation of the number of fluorescent cells in each original plate well during the subsequent analysis using a FACS Calibur system (BD Biosciences). Data acquisition and analysis of the results were performed using the CELLQuest software. The number of beads acquired was received by gating the bead population in an FL-1-Fl-2 dot plot. The cell population was gated in the Forward Scatter (FSC)–Side Scatter (SSC) dot plot, and 10 000 events within this gate were acquired from each sample. This was instead of excluding debris using a threshold, because that might exclude beads during acquisition. At least 10 000 events above the bead population, on the x-axis (FSC), excluding debris and TruCOUNT™ beads, were analyzed for each sample. Results were reported as the mean channel of fluorescence and as the proportion of cells that fluoresce.
The cell culture supernatants, from cells analyzed using flow cytometry, were analyzed for the level of TNF-α secreted during the last approximately 20 h using ELISA. Dead cells and debris were removed from supernatants by centrifugation at 510g for 5 min. The supernatants were then stored at −20°C until further use. ELISAs were performed using the Quantikine® Mouse TNF-α Immunoassay kit (R&D Systems, Minneapolis, MI, USA), according to the protocol provided by the manufacturer. Samples, controls and standards were assayed in duplicate. The optical density was determined within 30 min, using an Emax precision microplate reader (Molecular Devices, Sunnyvale, CA, USA) set to 450 nm, with wavelength correction set to 562 nm. The sample concentrations were calculated by the SOFTmax PRO, version 3.1 software (Molecular Devices).
For analysis of the localization of the I/St and A/Sn TNF-α, transfected cells were fixed attached to the coverslips, with 3% PFA and 5% sucrose for 30 min, at time points 48 and 72 h post-transfection. Wells were washed two times with 500 μl PBS. Fixed cells were stained with Vybrant™ DiI (Molecular Probes), for visualization of the cell membrane, by incubation in 37°C for 15 min, followed by washing three times with 500 μl PBS, and DAPI-staining (0.5 μg/ml in PBS; Molecular probes, Eugene, OR, USA) for 15 min. Cells were washed again, twice with 500 μl PBS. The coverslips were mounted onto glass slides using polyvinyl alcohol mounting medium with DABCO™ (antifading) (Fluka Chemie, Sigma-Aldrich). Slides were stored at −20°C until further use. Analysis was performed using an LSM 510 META confocal microscopy (Zeiss, Oberkochen). For analysis of the DiI, DAPI and ZsGreen, an He/Ne1, UV and Argon laser, respectively, was used. Pictures were created using a mean of eight scans, using the LSM 510 software (Zeiss).
A TNF-α secretion variable, corrected for differences in cell number between samples, was constructed by dividing the TNF-α concentration in the supernatant with the ratio:number of TNF-α-ZsGreen producing cells in the plate well (based on the proportion of cells that fluoresce and the total cell number, both monitored by flow cytometry)/the average cell number per well. Intracellular level of TNF-α-ZsGreen was defined as the mean channel of fluorescence determined by flow cytometry. A variable based on the secretion corrected for differences in intracellular TNF-α-ZsGreen level was constructed, by dividing the secretion variable with the ratio: mean channel of fluorescence from the TNF-α-ZsGreen producing cells/the average mean fluorescence channel number. Secretion and intracellular level of TNF-α-ZsGreen were analyzed for dependence on protein variant and post-transfection time point. Variance equality between protein variants and post-transfection time points was assessed using Bartlett's test. Univariate multiple regression and two-way ANOVA were performed. Dependence on protein variant was then tested at each time point using two-tailed t-tests. A P-value of less than 0.05 was considered significant. A significant effect of protein variant on the secretion variable, motivated analysis of the underlying parameters proportion and number of TNF-α-ZsGreen producing (ie fluorescing) cells. The analyses were calculated using STATA version 6.0 (Stata Corporation, College Station, TX, USA).
Nikonenko BV, Apt AS, Moroz AM, Averbakh MM . Genetic analysis of susceptibility of mice to H37Rv tuberculosis infection: sensitivity versus relative resistance. Prog Leuk Biol 1985; 3: 291–298.
Lavebratt C, Apt AS, Nikonenko BV, Schalling M, Schurr E . Severity of tuberculosis in mice is linked to distal chromosome 3 and proximal chromosome 9. J Infect Dis 1999; 180: 150–155.
Nikonenko BV, Apt AS, Moroz AM, Averbakh MM . Comparative analysis of mycobacterial infections in susceptible I/St and resistant A/Sn mice. Tuberc Lung Dis 2000; 80: 15–25.
Sanchez F, Radaeva TV, Nikonenko BV et al. Multigenic control of disease severity after virulent Mycobacterium tuberculosis infection in mice. Infect Immun 2003; 71: 126–131.
Dinarello CA . Proinflammatory cytokines. Chest 2000; 118: 503–508.
Vassalli P . The pathophysiology of tumor necrosis factors. Annu Rev lmmunol 1992; 10: 411–452.
Sharma R, Anker SD . Cytokines, apoptosis and cachexia: the potential for TNF antagonism. Int J Cardiol 2002; 85: 161–171.
Ware CF, VanArsdale S, VanArsdale TL . Apoptosis mediated by the TNF-related cytokine and receptor families. J Cell Biochem 1996; 60: 47–55.
Dinarello CA . Proinflammatory and anti-inflammatory cytokines as mediators in the pathogenesis of septic shock. Chest 1997; 112 (Suppl): 321S–329S.
Flynn JL . Immunology of tuberculosis and implications in vaccine development. Tuberculosis 2004; 84: 93–101.
Flynn JL, Goldstein MM, Chan J et al. Tumor necrosis factor-alpha is required in the protective immune response against Mycobacterium tuberculosis in mice. Immunity 1995; 2: 561–572.
Bean AG, Roach DR, Briscoe H et al. Structural deficiencies in granuloma formation in TNF gene-targeted mice underlie the heightened susceptibility to aerosol Mycobacterium tuberculosis infection, which is not compensated for by lymphotoxin. J Immunol 1999; 162: 3504–3511.
Roach DR, Bean AG, Demangel C, France MP, Briscoe H, Britton WJ . TNF regulates chemokine induction essential for cell recruitment, granuloma formation, and clearance of mycobacterial infection. J Immunol 2002; 168: 4620–4627.
Keane J, Gershon S, Wise RP et al. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent. N Engl J Med 2001; 345: 1098–1104.
Knight J . Polymorphisms in tumor necrosis factor and other cytokines as risks for infectious diseases and the septic syndrome. Curr Infect Dis Rep 2001; 3: 427–439.
Hajeer AH, Hutchinson IV . Influence of TNFalpha gene polymorphisms on TNFalpha production and disease. Hum Immunol 2001; 62: 1191–1199.
Tang P, Hung MC, Klostergaard J . Human pro-tumor necrosis factor is a homotrimer. Biochemistry 1996; 35: 8216–8225.
Kriegler M, Perez C, DeFay K, Albert I, Lu SD . A novel form of TNF/Cachectin is a cell surface cytotoxic transmembrane protein: ramifications for the complex physiology of TNF. Cell 1988; 53: 45–53.
Black RA, Rauch CT, Kozlosky CJ et al. A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells. Nature 1997; 385: 729–733.
Moss ML, Jin SL, Milla ME et al. Cloning of a disintegrin metalloproteinase that processes precursor tumour-necrosis factor-alpha. Nature 1997; 385: 733–736.
Perez C, Albert I, DeFay K et al. A nonsecretable cell surface mutant of tumor necrosis factor (TNF) kills by cell-to-cell contact. Cell 1990; 63: 251–258.
Decoster E, Vanhaesebroeck B, Vandenabeele P et al. Generation and biological characterization of membrane-bound, uncleavable murine tumor necrosis factor. J Biol Chem 1995; 270: 18473–18478.
Grell M, Douni E, Wajant H et al. The transmembrane form of tumor necrosis factor is the prime activating ligand of the 80 kDa tumor necrosis factor receptor. Cell 1995; 83: 793–802.
Utsumi T, Takeshige T, Tanaka K et al. Transmembrane TNF (pro-TNF) is palmitoylated. FEBS Lett 2001; 500: 1–6.
Ruuls SR, Hoek RM, Ngo VN et al. Membrane-bound TNF supports secondary lymphoid organ structure but is subservient to secreted TNF in driving autoimmune inflammation. Immunity 2001; 15: 533–543.
Majorov KB, Lyadova IV, Kondratieva TK et al. Different innate ability of I/St and A/Sn mice to combat virulent Mycobacterium tuberculosis: phenotypes expressed in lung and extrapulmonary macrophages. Infect Immun 2003; 71: 697–707.
Eruslanov EB, Majorov KB, Orlova MO et al. Lung cell responses to M. tuberculosis in genetically susceptible and resistant mice following intratracheal challenge. Clin Exp Immunol 2004; 135: 19–28.
Bastos KR, Marinho CR, Barboza R, Russo M, Alvarez JM, D’Imperio Lima MR . What kind of message does IL-12/IL-23 bring to macrophages and dendritic cells? Microbes Infect 2004; 6: 630–636.
Xing Z, Zganiacz A, Santosuosso M . Role of IL-12 in macrophage activation during intracellular infection: IL-12 and mycobacteria synergistically release TNF-alpha and nitric oxide from macrophages via IFN-gamma induction. J Leukocyte Biol 2000; 68: 897–902.
Bekker LG, Moreira AL, Bergtold A, Freeman S, Ryffel B, Kaplan G . Immunopathologic effects of tumor necrosis factor alpha in murine mycobacterial infection are dose dependent. Infect Immun 2000; 68: 6954–6961.
Alexopoulou L, Pasparakis M, Kollias G . A murine transmembrane tumor necrosis factor (TNF) transgene induces arthritis by cooperative p55/p75 TNF receptor signaling. Eur J Immunol 1997; 27: 2588–2592.
Tracey KJ, Fong Y, Hesse DG et al. Anti-cachectin/TNF monoclonal antibodies prevent septic shock during lethal bacteraemia. Nature 1987; 330: 662–664.
Mohler KM, Sleath PR, Fitzner JN et al. Protection against a lethal dose of endotoxin by an inhibitor of tumour necrosis factor processing. Nature 1994; 370: 218–220.
Mueller C, Corazza N, Trachsel-Loseth S et al. Noncleavable transmembrane mouse tumor necrosis factor-alpha (TNFalpha) mediates effects distinct from those of wild-type TNFalpha in vitro and in vivo. J Biol Chem 1999; 274: 38112–38118.
Zheng Y, Saftig P, Hartmann D, Blobel C . Evaluation of the contribution of different ADAMs to tumor necrosis factor α (TNFα) shedding and of the function of the TNFα ectodomain in ensuring selective stimulated shedding by the TNFα convertase (TACE/ADAM17). J Biol Chem 2004; 279: 42898–42906.
Gearing AJ, Beckett P, Christodoulou M et al. Processing of tumour necrosis factor-alpha precursor by metalloproteinases. Nature 1994; 370: 555–557.
Solomon KA, Covington MB, DeCicco CP, Newton RC . The fate of pro-TNF-alpha following inhibition of metalloprotease-dependent processing to soluble TNF-alpha in human monocytes. J Immunol 1997; 159: 4524–4531.
Glaser KB, Falduto M, Metzger R et al. Expression, release, and regulation of human TNF from stable transfectants of HEK-293 cells. Inflamm Res 1997; 46 (Suppl. 2): S127.
Solomon KA, Pesti N, Wu G, Newton RC . Cutting edge: a dominant negative form of TNF-α converting enzyme inhibits proTNF and TNFRII secretion. J Immunol 1999; 163: 4105–4108.
Reddy P, Slack JL, Davis R et al. Functional analysis of the domain structure of tumor necrosis factor-α converting enzyme. J Biol Chem 2000; 275: 14608–14614.
Itai T, Tanaka M, Nagata S . Processing of tumor necrosis factor by the membrane-bound TNF-α-converting enzyme, but not its truncated soluble form. Eur J Biochem 2001; 268: 2074–2082.
Utsumi T, Akimaru K, Kawabata Z et al. Human pro-tumor necrosis factor: molecular determinants of membrane translocation, sorting, and maturation. Mol Cell Biol 1995; 15: 6398–6405.
Cseh K, Beutler B . Alternative cleavage of the cachectin/tumor necrosis factor propeptide results in a larger, inactive form of secreted protein. J Biol Chem 1989; 264: 16256–16260.
Utsumi T, Levitan A, Hung MC, Klostergaard J . Effects of truncation of human pro-tumor necrosis factor transmembrane domain on cellular targeting. J Biol Chem 1993; 268: 9511–9516.
Han J, Brown T, Beutler B . Endotoxin-responsive sequences control cachectin/tumor necrosis factor biosynthesis at the translational level. J Exp Med 1990; 171: 465–475.
Han J, Beutler B . The essential role of the UA-rich sequence in endotoxin-induced cachectin/TNF synthesis. Eur Cytokine Netw 1990; 1: 71–75.
Beutler B . Regulation of cachectin biosynthesis occurs at multiple levels. Prog Clin Biol Res 1990; 349: 229–240.
Hel Z, Skamene E, Radzioch D . Two distinct regions in the 3′ untranslated region of tumor necrosis factor alpha mRNA form complexes with macrophage proteins. Mol Cell Biol 1996; 16: 5579–5590.
Caput D, Beutler B, Hartog K, Thayer R, Brown-Shimer S, Cerami A . Identification of a common nucleotide sequence in the 3′-untranslated region of mRNA molecules specifying inflammatory mediators. Biochemistry 1986; 83: 1670–1674.
Han J, Huez G, Beutler B . Interactive effects of the tumor necrosis factor promoter and 3′-untranslated regions. J Immunol 1991; 146: 1843–1848.
Schug J, Overton GC . Technical Report CBIL-TR-1997-1001-v0.0. Computational Biology and Informatics Laboratory School of Medicine, University of Pennsylvania, Pennsylvania, 1997.
This study was supported by grants from Swedish Medical Research Council, Karolinska Institutet Foundation, Magnus Bergwall Foundation, and Swedish Heart and Lung Foundation. We thank Ricardo Giscombe for skilful assistance with the flow cytometry experiments.
About this article
Cite this article
Kähler, A., Persson, AS., Sánchez, F. et al. A new coding mutation in the Tnf-α leader sequence in tuberculosis-sensitive I/St mice causes higher secretion levels of soluble TNF-α. Genes Immun 6, 620–627 (2005). https://doi.org/10.1038/sj.gene.6364249
- leader peptide
Mouse models of human TB pathology: roles in the analysis of necrosis and the development of host-directed therapies
Seminars in Immunopathology (2016)
Seminars in Immunology (2014)
The Journal of Pathology (2013)
A locus involved in tuberculosis infection control in mice locates in the proximal part of the H2 complex
Molecular Biology (2011)
Molecular Biology (2008)