Tolerating Aire - Out of the Thymus

The immune system protects us against infections but it must also tolerate the many different proteins that constitute our bodies. T lymphocytes of the immune system that react against the body’s own proteins could trigger autoimmunity, so they are usually deleted during their maturation in the thymus. A protein called Aire (Autoimmune regulator) turns on expression in the thymus of nearly 2,000 genes that are otherwise expressed in only a few organs and tissues, thereby exposing new T cells to these proteins and avoiding autoimmune disease. Indeed, mutations in Aire result in a broad autoimmunity against many organs and tissues. Previously, Aire was thought by many to be expressed only in the thymus.

Gardner and colleagues detected Aire in rare stromal (non-lymphoid) cells they called extrathymic Aire-expressing cells (eTACs) within the spleen, lymph nodes, and Peyer’s patches (nodes associated with the gut). These cells are located between the T and B cell areas of these organs but are highly mobile. Surprisingly, Aire turned on different genes in eTACs than were turned on in the thymus, suggesting that Aire expression in eTACs is not a ‘safety net’ for T cells that accidentally escaped the thymus but rather acts to delete additional autoreactive T cells. In eTACS, Aire turned on fewer than 200 genes, but several of these are suspected to be involved in autoimmunity, including desmoglein 1a (in pemphigous foliaceus) and the N-methyl-D-aspartate receptor 2C (in lupus). In an animal model of type 1 diabetes, eTACs expressing glucose-6-phosphate-related protein deleted T cells specific for this protein and prevented onset of this autoimmune disease (Fig. 2f shown, open squares represent mice expressing glucose-6-phosphatase-related Adig in eTACs).

These findings suggest that manipulating eTACs is a promising approach to therapy for autoimmune disease. Comparing eTACs in healthy people and autoimmune patients might also help identify deficiencies that can be corrected to prevent or ameliorate disease.

Gardner JM, Devoss JJ, Friedman RS, Wong DJ, Tan YX, Zhou X, Johannes KP, Su MA, Chang HY, Krummel MF, Anderson MS. Deletional tolerance mediated by extrathymic Aire-expressing cells. Science. 2008 Aug 8;321(5890):843-7.

Homozygosity mapping: Autism in the family

Most genes identified in genome-wide analyses contribute only modestly to disease risk or protection, with already low relative risks changed only ~20%, and taken together probably contribute to a minority of disease cases. Morrow and colleagues took a different approach, believing that analyzing individual families might reveal higher-risk genes for autism spectrum disorders (ASDs). They noted that offspring of first-cousins have twice as many neurological birth defects, including ASDs.

They also noted that a significant involvement of autosomal recessive genes would be 'signaled' by change in the male-to-female ratio (because the M:F ratio is typically 1:1 for autosomal recessive traits vs. male-dominant ASD). Indeed, among the 104 families recruited for the study, the M:F ratio was 2.6 for the offspring of the 88 consanguineous families (cousins) vs. 7.4 for the non-consanguineous.

They genotyped using arrays of SNPs and BACs, the latter validated for comparative genomic hybridization (CGH) detection of copy number variation (CNV) due to deletions, etc. Few de novo (not inherited) CNVs were detected. Large deletions on both DNA strands of the affected child (homozygous), and one strand of each parent (hemizygous), were found in 5 of the 78 consanguineous families but not among any of the other ~400 ASD cases or ~2,400 controls. In the figure (from Figure 1C), the SNPs spanning a large deletion in the AU-3100 pedigree is shown: the parents 3103 & 3104 (top and second) are hemizygous as is one child (3102, third down) but the other child, 3101 (bottom), who is homozygous for the deleted chromosome, has autism and seizures.

The investigators were surprised to find 3 genes linked to the 2 largest deletions were previously identified as being regulated by neural activity, and thus candidates for involvement with learning. Mutations in one of these genes, NHE9, were detected in non-consanguineous families and associated with additional neurological disorders (epilepsy) and learning delays (language acquisition).

EM Morrow et al. Identifying autism loci and genes by tracing recent shared ancestry. Science. 2008 Jul 11;321(5886):218-23.

Therapeutic Interruption of Protein Degradation

Some antibodies that bind ‘self’ molecules (autoantibodies) can signal and even cause autoimmune diseases. Antibodies are produced by B cells, largely by a differentiated, typically short-lived variety called plasma cells (PC), which dedicate 10-20% of their protein-synthesis capacity to immunoglobulin. PC develop in the germinal centers of lymph nodes and spleen and then migrate to the bone marrow, where a subset endures. Treatments that target B cells, such as Rituxan, might often fail to substantially reduce autoantibodies because PC are resistant.

Neubert and colleagues reasoned that precisely the hallmark of PC – their high rate of protein synthesis – could sensitize them to bortezomib (Bz, Velcade), a proteasome inhibitor that is therapeutic in some cancers. Bz interferes with the ubiquitin protein degradation pathway, thereby blocking NF-kB release and promoting the unfolded protein ‘stress’ response that induces apoptosis. Here, they show that Bz reduced short- and long-lived PC (CD138+ CD25- cells with cytoplasmic immunoglobulin light or heavy chains) ~90% in the bone marrow and spleens of mice after just 48 h treatment. Antibody-secreting cells remained decreased during 8 weeks of treatment. Bz had little or no effect on total B cell numbers or many B cell subsets, although germinal center B cells were reduced (Fig. 1). Cyclophosphamide or dexamethasone were less effective in reducing the numbers of long-lived (BrdU-) PC, suggesting a cellular basis for the failure of these current therapeutics in reducing autoantibodies (Fig. 2). In bone marrow and splenic PC, Bz rapidly induces transient expression of CHOP (20-40 fold within 4 h), a signaling protein previously implicated in the apoptotic response to the stress of protease inhibitors.

Lastly, they treated lupus-prone mice (NZB/W F1) with Bz and found it prevented onset of kidney disease (measured by proteinuria) and death. Remarkably, Bz was therapeutic in early disease, reducing serum levels of the autoantibody associated with kidney disease and preventing proteinuria in this lupus model as well as another (MRL/lpr: shown, dotted line Bz treated, solid line control treated, from Fig. 5a).

As Bz is already approved for use in humans, it should not be long before we know its therapeutic efficacy in lupus and other autoimmune diseases. Also, a new generation of proteasome inihibitors that target specific steps in the process are ready to be tested.

Neubert K, Meister S, Moser K, Weisel F, Maseda D, Amann K, Wiethe C, Winkler TH, Kalden JR, Manz RA, Voll RE. “The proteasome inhibitor bortezomib depletes plasma cells and protects mice with lupus-like disease from nephritis” Nat Med. 2008 Jun 8.

Adjuvants: a little less ‘dirty’, less secret

How do adjuvants – mixtures that promote immunization/vaccination – work? Malherbe and colleagues applied modern tools and ideas to answer this old question. They analyzed the immune responses of mice (B10.BR strain) to the protein PCC (pigeon cytochrome C), which they had previously shown to be dominated by helper T cells (Th) responding to a single PCC fragment (peptide ‘epitope’ associated with I-Ek) and expressing T cell receptors (TCR) encoded by Valpha11Vbeta3 (a11b3). They concluded from those previous studies that TCR affinity, above a threshold level, does not drive clonal diversity.

Here, they compared Th numbers and diversity after subcutaneous immunization in an aluminum precipitate (Alum), or in an emulsion known as “incomplete Freund’s adjuvant” (IFA), or complete Freund’s adjuvant (CFA = IFA + dead mycobacterium), or with aqueous oligodeoxynucleotide (CpG) that stimulates the innate immunity Toll Like Receptor-9 (TLR-9), or monophosphoryl lipid A (MPL) that stimulates TLR-4.

First, they compared the numbers of TCRa11b3+ Th in draining lymph nodes on day 7 after immunization with or without PCC antigen and found a ~30 fold range among the adjuvants. MPL induced the largest number of PCC-specific Th but also many nonspecific Th. CpG induced the largest differential between PCC-specific and nonspecific Th whereas Alum induced nearly as many nonspecific as specific Th. The response kinetics were all similar, peaking around day 7. Second, they analyzed the TCR sequence “features…that assort with PCC-specificity” in dozens of single Th cells, concluding that clonal dominance occurred with or without TLR agonists or antigen ‘depots’ (IFA, CFA, and Alum). Third, they found lower affinity TCR Jbeta2.5 regions predominated among Th responding to depot-forming adjuvants, whereas higher affinity Jbeta1.2 predominated among Th responding to the non-depot forming adjuvants (CpG and MPL). Also, Vbeta usage depended upon the adjuvant.
Fourth, in affinity tests using PCC-I-Ek tetramers, they found a range of responses similar to those they had found in TCRa11b3+ cells (from Figure 4C, open circles are immunizations without PCC, bars are immunizations with PCC). Also consistent with the sequence analysis, CpG and MPL induced higher levels of binding as measured by mean fluorescence intensity (MFI).
Fifth, upon PCC immunization of mice hosting Th with transgenic high- or low-affinity TCRs, they found that IFA stimulated both populations equally whereas MPL stimulated Th with the high-affinity TCR but not those with the low-affinity TCR. Sixth, they found MPL as effective as the depot-forming IFA in activating transgenic Th transferred 5 days after immunization. Finally, 400 ug PCC in MPL induced as much Th proliferation as did 40 ug without changing the ratio of high- and low-affinity TCR, though 4 ug was suboptimal.

These parameters will be useful guides in determining how adjuvants improve immune protection, the goal of vaccination, especially in humans.

Malherbe L, Mark L, Fazilleau N, McHeyzer-Williams LJ, McHeyzer-Williams MG. “Vaccine adjuvants alter TCR-based selection thresholds”. Immunity. 2008 May;28(5):698-709.

The 3% (sialylated) Solution

Intravenous Immunoglobulin (IVIG) is an immune modulator and anti-inflammatory agent used to treat a wide range of diseases, including rheumatoid arthritis (see also: 1990 NIH consensus panel report). IVIG contains antibodies (immunoglobulins, Ig) pooled from the blood of hundreds or even thousands of donors; pooling raises the risks of infection and complicates standardization.

Here, Anthony and colleagues tested their idea that sialylation is required for the therapeutic activity of the invariant region of Ig called Fc (fragment crystallizable). Sialic acid can be linked through alpha 2,3 or alpha 2,6, which are distinguished by the location of the hydroxyl and methyl groups on the 6 member ring.
First, they looked for these linkages in a preparation of IVIG by comparing its mass spectrum with standards. Nearly all the sialylated IVIG was alpha 2,6 linked and treatment with an enzyme that selectively removed this linkage abolished the anti-inflammatory activity of IVIG. Of potentially enormous significance was their success in making Fc derived from cloned human IgG therapeutic by alpha 2,6 sialylation. The figure shows the clinical score of mice made arthritic by administration of K/BxN sera alone (top line), or also treated therapeutically with IVIG (second line), 1/30 as much sialylated-enriched Fc from IVIG (0.033 g/kg SNA IVIG, third line), or an equivalent amount of recombinant, sialylated Fc (bottom line). This finding could lead directly to the manufacture of much safer, much more consistent IVIG preparations.
Anthony RM, Nimmerjahn F, Ashline DJ, Reinhold VN, Paulson JC, Ravetch JV. Science. 2008 Apr 18;320(5874):373-6. "Recapitulation of IVIG anti-inflammatory activity with a recombinant IgG Fc".

Microbe cheats death by misdirecting killer enzyme chaperone

Two million people die each year from tuberculosis and about a third of all humans are infected with the intracellular parasite responsible: Mycobacterium tuberculosis. This microbe overcomes the defenses of the macrophages it infects by inhibiting the fusion of phagosomes with lysosomes and avoiding the activity of reactive oxygen and nitrogen intermediates, including nitric oxide (NO). NO is extremely reactive and consequently short-lived, so the microbe misdirects the enzyme that produces NO, inducible NO synthase (iNOS), away from the phagosome. Davis and colleagues guessed that EBP50, a protein previously implicated in anchoring iNOS and other cellular proteins to the actin cytoskeleton, might be responsible.

They first showed that EBP50 is colocated with iNOS in the phagosomes of activated macrophages but less so after infection by M. tuberculosis. Reducing EBP50 with siRNA also reduced colocalization. Moreover, siRNA knockdown of EBP50 increased the viability of M. tuberculosis in infected macrophages, demonstrating its involvement in controlling this microbe (Figure, coronin is a control cytoskeletal chaperone that does not bind EBP50). However, since EBP50 is involved in many cellular processes, the specificity of the inhibition remains unproven.

Davis AS, Vergne I, Master SS, Kyei GB, Chua J, Deretic V. “Mechanism of Inducible Nitric Oxide Synthase Exclusion from Mycobacterial Phagosomes”. PLoS Pathog. 2007 Dec 7;3(12):e186

Identifying Friend or Foe - Mreg?

How do the immune cells within the gut distinguish between potentially dangerous pathogens and harmless or beneficial commensal organisms? [Hint: It’s not through PAMPs (pathogen-associated molecular patterns.)]

These investigators identified in mice a novel population of macrophages within the lamina propria, the layer of mucosa beneath the epithelium on the lumenal side of the intestine wall. These macrophages secrete IL-10 and other anti-inflammatory cytokines but no pro-inflammatory cytokines (even when stimulated through their Toll-like receptors by PAMPs). They also stimulate the development of regulatory T cells (Tregs) more potently than splenic macrophages (figure, CD4+ FoxP3+ Treg) and suppress the secretion of IL-17 by DC within the lamina propria. The authors suggest that a “dynamic interaction between these subsets may influence the balance between immune activation and tolerance”. They need to abbreviate the idea... how about 'Mregs'?

Denning et al. "Lamina propria macrophages and dendritic cells differentially induce regulatory and interleukin 17-producing T cell responses". Nat Immunol. 2007 Oct;8(10):1086-94.

Targeting the Plant Inside

Toxoplasm gondii are intracellular parasites that cause toxoplasmosis, the “third leading cause of death attributed to foodborne illness in the United States. More that 60 million [Americans] carry the Toxoplasma parasite, but very few have symptoms because the immune system usually keeps the parasite from causing illness” (CDC). T.gondii contain organelles or plastids called “apicoplasts” that were probably acquired in evolution along with an algal endosymbiont.

Apicoplasts contain unique proteins that may be effectively targeted by therapeutic agents. These researchers previously demonstrated a calcium control of protein secretion and identified conserved Ca-responsive proteins. Here they show that calcium-dependent development in T. gondii is controlled by a plant hormone produced by the apicoplast, abscisic acid (ABA, named for its role in abscission, the shedding of plant leaves, fruit, etc.). They found that fluridone, an herbicide that specifically inhibits an enzyme on the synthetic pathway producing ABA, blocked T gondii maturation. Finally, they demonstrated that the herbicide also worked as a therapeutic, protecting mice from a lethal dose of T gondii (shown, from Figure 4). Another infamous apicoplast-containing parasite is Plasmodium falciparum, which causes severe malaria, suggesting an exciting new approach to treating this scourge.

Nagamune et al. "Abscisic acid controls calcium-dependent egress and development in Toxoplasma gondii" Nature. 2008 Jan 10;451(7175):207-10.

Protracted IKKb inhibition is dangerous

Many pro-inflammatory molecules share the intracellular signaling pathway with Nuclear Factor – kappa Binding (NF-kB) for their induction and activity. Cytoplasmic NF-kB is released and transported into the nucleus, where it drives transcription of specific genes, when an inhibitor (IkB) is phosphorylated by the inhibitor-kappa kinase beta (IKKb) The importance of this pathway suggests that IKKb would be a promising target for new anti-inflammatory drugs. Here, Greten and colleages tested this idea using mice in which IKKb was deleted in myeloid cells: macrophages and neutrophils (IKKb-del-myo). [A total IKKb deficiency is an embryonic lethal]. IKKb-del-myo mice were viable and had no obvious changes even within tissues that are rich in myeloid cells, such as lymphoid organs and the gastrointestinal tract. They observed that IKKb-del-myo mice were actually more susceptible to endotoxin- (LPS-) induced toxic shock, which is mediated by TNF and interleukin-1beta (IL-1b). IKKb-deficient cells are more susceptible to apoptosis because some NF-kB-induced genes are protective (e.g., A20).

The authors also studied another conditional knockout mouse, in which IKKb is deleted in all “interferon-inducible” cells (IKKb-del), though this model is complicated by a requisite induction with the pleiotropic agent poly-I:C and deletion probably occurs in many more cell types. These mice develop many more granulocytes and, like the IKKb-del-myo mice, also produced much higher levels of IL-1b after LPS treatment. However, mRNA encoding IL-1 was actually decreased, suggesting a change in IL-1 processing.

Indeed, the authors determined that NF-kB inhibits caspase-1 and serine proteases that are required in macrophages and neutrophils, respectively, to cleave pro-IL-1b and release the active cytokine. Consistent with this interpretation, a serine protease inhibitor protected mice against IL-1b-induced death. Mice exposed to LPS die more quickly and more often when fed ML120B, an inhibitor of IKKb for a protracted period with multiple doses, correlating with an increased processing and release of IL-1b (figure, extracted from Figure 7). Definitely not a lead compound for a daily regimen.
Greten et al. Cell. 2007 Sep 7;130(5):918-31. "NF-kappaB is a negative regulator of IL-1beta secretion as revealed by genetic and pharmacological inhibition of IKKbeta"

Fatter and Sicker: Obesity weakens Immunity

This paper reminds me of an old joke: when a patient, having received a dubious diagnosis, asks for a second opinion, the doctor replies 'you're also ugly'. But seriously, folks, this paper suggests that the rise of obesity in the US could have unanticipated consequences for infectious disease. To test the effect of obesity on immunity, Amar and colleagues infected fat or lean mice with Porphyromonas gingivalis, a bacterium that causes gum disease. Mice fed with high fat food for 4 months weighed about 30% more than the mice on normal chow.
They found that mice with diet-induced obesity (DIO) suffered a greater loss of alveolar (gum) bone than did lean mice. P. gingivalis infection induced less inflammatory cytokine release in DIO mice than in lean mice. Macrophages cultured from DIO mice showed significantly reduced activation by LPS of the intracellular signaling molecule NF-kappaB (a chromosome immunoprecipitation is shown, from Fig. 8). Oddly, C-reactive protein, a marker of inflammation, is reported elevated in the bloodstream of obese humans. Do you have a 'second opinion'?
Amar et al., Diet-induced obesity in mice causes changes in immune responses and bone loss manifested by bacterial challenge. Proc Natl Acad Sci U S A. 2007 Dec 18;104(51):20466-71.