A new biomarker for assessing resistance to therapy in rheumatoid arthritis

According to a recently published study, low expression of CD39 on regulatory T cells (Tregs) could be used as a biomarker for resistance to methotrexate (MTX) therapy in rheumatoid arthritis (RA). MTX, an antimetabolic drug, is the first line of therapy for RA patients. However, up to 40% of patients are unresponsive to MTX (or UR-MTX resistant). There is a need to early diagnose MTX resistance as the resistance generally appears after 3 months of therapy and that means worsening of RA symptoms during that time. The study published in January issue of PNAS shows that low expression of CD39 on circulating Tregs is a rapid, reliable and convenient biomarker for MTX resistance.

RA is an inflammatory autoimmune disease associated with joint destruction and severe morbidity. MTX has been generally used as drug to treat RA. MTX has two known functions: inhibitor of dihydrofolate reductase/folic acid metabolism and reducing inflammation by elevation of extracellular adenosine (ADO). Tregs are regulatory T cells responsible for keeping the immune system in check by suppressive activation and proliferation of wide variety of cell types. One of the mechanisms through which Treg suppress other cells is through production of extracellular ADO by CD39/CD73, expressed on the surface of Tregs.

Given that the MTX also exerts its effect through generation of ADO by CD39/CD73 and that CD39/CD73 are also important in Tregs activity, the authors investigated possible link between MTX unresponsiveness and expression of CD39 on Tregs in RA patients. The authors first observed that parameters related to disease severity were significantly higher in UR-MTX patients than in R-MTX (responsive) patients. Furthermore, the number of circulating Tregs was lower in UR-MTX than in R-MTX. Encouraged by these findings, the researchers compared expression of CD39 and CD73 on Tregs from R-MTX and UR-MTX RA patients. The expression of CD39 on Trges from UR-MTX RA patients was significantly lower than that in R-MTX patients. To find out if the increase in number of CD39+Tregs in R-MTX was a result of MTX treatment, the authors followed some patients and studied CD39+ expression on T regs before and after treatment. Although there was no difference in percentage of CD39+ tregs in R and UR-MTX before treatment, their numbers increased much more in R-MTX patients than in UR-MTX patients following treatment. They further observed that the density of expression of CD39 was low in UR-MTX patients before and after treatment in comparison to R-MTX patients.

Various attempts have been made in the past to identify a biomarker for MTX unresponsiveness. This study provides a unique non-invasive approach of using CD39 expression on Tregs as biomarker for unresponsiveness.

How do vaccines interact with human cells?

In a recently published study, scientists have identified various transcriptional programs induced in human antigen presenting cells (APCs) by various vaccines. The article published in October issue of Nature Communications sheds new light on the mechanisms of vaccine interaction with human dendritic cells (DCs), a primary APC involved in vaccination response.

Previous studies have shown that vaccination mostly depend on dendritic cells. DCs are highly efficient in presentation of antigen. In response to a pathogen, DCs transcribe various set of genes that then interact with each other. Although, many sub-populations of DCs have been identified in human blood, skin and other tissues, the role of these populations in eliciting specific immune response to different vaccines is not entirely known. Furthermore, it is not much known about the various transcriptional programs initiated in response to different vaccines.

Romain Banchereau and colleagues from Baylor Institute for Immunology Research, Dallas and Jackson Laboratory for Genomic Medicine, Farmington used a unique multi-step approach to understand how human DCs respond to vaccine challenge in vitro. They studied transcriptional changes in DCs that were first generated by culturing monocytes (a precursor to DC) with different cytokines and then exposing them to a number of pathogens. The researchers observed a broad spectrum of unique and common transcriptional responses to pathogens over time. To further understand the biological significance of these transcriptional responses, the scientists generated a framework that groups transcripts into modules based on their co-expression across pathogens, time points, and DCs. They found that modules represented alteration of many biological pathways including interferon response, inflammation, antigen processing and presentation, DC maturation and T cell activation.

The investigators then applied the same approach to study transcriptional responses of various APCs to 13 commercially available vaccines in vitro and observed that response to different vaccine was mediated through unique APC subsets. In a further effort to understand the differences in immune pathways required for vaccination to those that are responsible for pathogenesis of a particular disease, the scientists applied the same framework to analyze transcriptional profile of individuals vaccinated with influenza vaccine and individuals with asymptomatic and symptomatic influenza infections. They found that vaccinated individuals showed similar transcriptional profiles to asymptomatic individuals and both showed a mild signature. In contrast, symptomatic individuals showed strong response. The researchers acknowledge that this approach can help in identifying which pathways are require for vaccination and which for establishing infection and this distinction between two pathways can help in development of safer and more effective vaccines in future.

Overall, this study provides a good understanding of transcriptional response of DCs to current vaccines and will certainly add in development of next generation vaccines, in which specific sub-populations of APCs can be targeted for a more effective response to a particular pathogen.

LL37 identified as an auto antigen in Psoriasis

According a new study published in nature communications, anti-microbial peptide (AMP) LL37 acts as an auto antigen in psoriasis. The study shows that LL37 was recognized as an antigen by circulating T cells in 46% of psoriatic patients. This recognition was shown by proliferation of T cells and production of interferon gamma in response to LL37. None of the controls showed such response.

Roberto Lande and colleagues from Italy and Switzerland published their findings in December issue of Nature communications. They performed their research using blood from psoriasis patients.

Psoriasis is a common condition of skin characterized by development of red scaly plaques on skin. Although it is sometimes considered an autoimmune disease, the nature of the auto antigen that leads to development of auto reactive T cells is not known. LL37 is an AMP that has been shown to be overexpressed in psoriatic skin and play important roles in pathogenesis of psoriasis. It forms complexes with extracellular self-nucleic acid. These complexes then activate plasmatoid and myeloid DCs which in turn leads to activation of T helper cells. This process then starts an inflammatory cascade. However, it was not known if LL37 can directly serve as an auto-antigen and stimulate T cells.

In this study, the authors established that psoriatic T cells were specifically responding to LL37 by showing that T cells respond only to LL37 when stimulated with a variety of other AMPs. The authors also looked at correlation of response to LL37 and disease status and found that up to 75% of psoriatic patients with moderate to severe plaque psoriasis and with a Psoriasis Activity Severity Index (PASI) score of more than 10 responded to LL37. The LL37 specific T cells in psoriasis patients showed production of pathogenic cytokines including IL17 as well as showed direct chemo tactic activity for inflammatory cells and cytotoxic ability. The study also showed that these T cells possess the ability to migrate into psoriatic skin lesions. This was further validated by finding of LL37 specific cell from skin biopsy of a psoriatic patient who earlier showed positive response of blood T cells to LL37.

In conclusion, this study shows that LL37 is an auto antigen in psoriasis and could be an appealing target to treat psoriasis.

Link between placental microbiome and pre-term birth

According to a recently published study in Science translation medicine, the placental microbiome might be associated with pre-term birth less than 37 weeks, and a remote history of antenatal infection in the first trimester. In this study, which was based on placental collection from 320 subjects, the investigators analyzed the human placental microbiome and provided details on microorganisms present, their probable functions and overall structure of placental microbial community.

Due to recent advances in sequencing technologies, a large number of studies have looked at the structure, function and diversity of human microbiome across multiple body sites in both healthy and disease states. Studies conducted on newborn babies have shown that the gut microbiome of neonates show complex microbial communities in the first few weeks of life. However, the source of these microbes and the time at which the infants acquire those is not yet known. It has also been shown that gut microbiome of full-term infants is different from that of low birth weight infants during the first few weeks of life, indicating that the infants might acquire these microbes in utero through a common source and that the composition of the microbiome might change with length of gestation. The authors of this study hypothesized that the human placenta could be that source.

The investigators of this study used whole-genome shotgun (WGS), a technique involving sequencing long strands of DNA, to determine the taxonomic classification of placental microbiome. They noted that several species of oral microbiome were detected in the placenta. These included Prevotella tannerae and non-pathogenic Neisseria species. They also found high abundance of E. coli in most individuals. The authors further compared the microbial community observed in placenta with those reported from other body sites (oral, stool, skin, nasal and vaginal) from non-pregnant healthy subjects using publicly available HMP (Human Microbiome Project) data. They observed that placental microbiota showed similarity to oral phyla only. Furthermore, the relative abundance of metabolic pathways was different in placenta when compared to other body sites. For example, the metabolism of co-factors and vitamins showed a greater relative abundance in placental functional gene profile.

The authors found a statistically significantly association between placental microbiome and pre-term birth less than 37 weeks. The researchers also identified various placental taxa whose abundance was enriched or diminished among women who experienced pre-term births. They also observed a statistically significant association between placental microbial community and a remote history of antepartum infection. Streptococcus and Acinetobacter were found to be enriched in women with a remote history of antepartum infection.

In conclusion, Aagaard and colleagues show that in contrast to the widespread believe that intra-uterine environment is sterile, human placenta harbors a variety of non-pathogenic commensal species, and the overall microbiome profile of placenta is most similar to that of non-pregnant human oral cavity. They also observed correlation between placental microbiome and pre-term birth and a history of antenatal infection.

The authors also noted several limitations of the study. The authors compared the placental microbiome with microbiome of other body sites from non-pregnant subjects. Moreover, the authors were not able to characterize the placental microbiome in early preterm gestation in women who had a full term delivery. Still, the study presents some interesting findings and indicates towards an essential role of placental microbiota in human pregnancy.

An interesting paper on HERVs

Expression of Human Endogenous Retrovirus Type K Envelope Protein is a Novel Candidate Prognostic Marker for Human Breast Cancer

I recently happened to read an interesting short report published in August 2011 issue of Genes and Cancer. In this study, the authors evaluated the use of expression of envelope (env) protein of human endogenous retroviruses-K (HERV-K) as a prognostic marker of breast cancer. HERVs constitute about 8% of human genome. There are many distinct families of HERVs present in human genome. Most of these families are defective and could not encode complete viruses. However, HERV-K family shows a conservation of seemingly intact retroviral genes as it is one of the most recently integrated viruses.   In a previously conducted study, the authors observed the expression of HERV-K env mRNA and protein in breast cancer tissues obtained from US women. To further study the role of HERV-K env mRNA as a prognostic marker for breast cancer, the authors looked at expression of HERV-K env mRNA and protein in breast cancer tissues obtained from both Chinese and US women. The authors also analyzed the relationship between HERV-K env expression and disease severity in these women.

Methodology and Results:

One hundred and twenty tissues from Chinese women including 40 breast cancer tissues, 40 matched adjacent uninvolved breast tissues and 40 breast tissues from benign breast diseases but not cancer were used in this study. HERV-K env mRNA was detected in 70% of breast cancer samples and 20% of surrounding uninvolved tissues from breast cancer patients. No HERV-K env mRNA was detected in breast tissues from control patients. Similar to mRNA results, HERV-K env protein expression was also higher in cancer tissues in comparison to controls. The authors also found medium to strong expression of env protein in cancerous tissues and weak to no expression in adjacent non-cancerous tissues.

The authors then looked at relationship between expression level of HERV-K protein and clinical features of the patients. 82% of the patients with high expression of HERV-K in their tumor had a tumor size greater than 3 cm. HERV-K env expression also correlated with other clinical and pathological characteristics in these patients. Furthermore, the scientists conducted a Kaplan-Meier survival analysis and found that the patients with high expression of HERV-K env protein had a significantly shorter overall survival in comparison with patients with moderate to low HERV-K env protein expression.

Taken together, all these results indicate that HERV-K env protein expression is highly upregulated in human breast cancer tissues and not in adjacent non-cancerous tissues or normal breast tissues.

In all the paper is nicely written and all the experiments seems to be nicely conducted. However, this paper exclusively talks about env but what about pol and gag? Expression of pol is particularly important as pol encodes for reverse transcriptase, which is required for production of new virus progenies.

The study did not observe any association between HERV-K env expression and ER/PR status. These results were surprising as earlier studies have observed that expression of HERV-K mRNA and protein was higher in breast cancer cell lines treated with estradiol and progesterone than in cells without treatment

Preterm infants' T cell responses to inactivated Poliovirus vaccine

I am writing a new blog after a long long time. Well, what a better way to start than this article published in 2009 issue of JID. http://jid.oxfordjournals.org/content/201/2/214.long

Background: Preterm infants are more susceptible to infection as their immune system is immature. earlier studies have examined humoral immune responses to many viral and bacterial vaccines in preterm infants. These studies have found disparity in antibody production between preterm and term infants. However, not much is known about T cell response to these vaccines in preterm infants. Since, adequate antibody production depends upon T cell help to B cells, the disparities in antibody production may reflect disparity in T cell responses in these two groups.
In the present study, the authors recruited 33 preterm and 50 term infants and studies their T cell response to vaccines.

Results: Of the 33 preterm infants, final results were available for 27 infants and of the 50 term infants, final results were available for 35 infants. First, the authors looked at the frequency of circulating CD4+ T cells in these two groups at 2 months and 7 months of age. There was no change in the frequency at these two time periods. Next, the authors looked at memory Cd4+ cells (Cd4+CD45RO+) in these infants. They observed that 7 month old preterm infants had fewer memory cells than seven month old term infants.
Furthermore, the authors stimulated CD4+ cells from both set of infants with Staphylococcus entertoxin B, Sabin Oral polio vaccine, serotype 3 strain and uninfected vero cell lysate by culturing whole blood with these in the presence of costimulatory monoclonal antibodies to CD28 and CD49d. Then they looked for percentage of CD4+ cells positive for CD69 and IFN-g. In response to stimulation with SEB, preterm infants had lower frequency of CD4+CD69+IFN-g+ cells than did term infants at 2 and seven months of age. The authors then compared the frequency of CD69+ and IFN-g+ cells among CD4 and CD45RO cells in both group of infants before and after vaccination with OPV serotype 3. Both the groups of infants developed a significant increase in poliovirus specific Cd69+IFg+ cells relative to prevaccination responses.
When the authors compared the CD69+ IFN-g+ responses between preterm and term infants after vaccination, they observed that frequency of CD4+CD69+, CD4+CD69+IFN-g+, CD4+CD45RO+CD69+ and CD4+CD45RO+CD69+IFN-g+ cells was comparable in two groups. To evaluate cell mediated immunity, PBMCs were isolated from both groups of infants (pre and post vaccination) and stimulated with OPV serotype 3. A positive proliferative response was defined as stimulation index of 3. It was observed that mean poliovirus-type 3 specific proliferation after vaccination was significantly lower in preterm infants than in term infants.
The authors next looked at neutralizing antibody response to all three subtypes of poliovirus in the sera of these infants after vaccination. All the infants showed neutralizing anitbodies post vaccination. There was no significant difference in geometric mean titer (GMT) to serotype 2 and 3 in the two groups of infants. However, GMT to serotype 1 was significantly low in preterm infants. To compare the safety of this vaccine in preterm and term infants, the authors looked for medically attended events (MAE) in both groups for 30 days. There were no significant events.

Discussion: The present study establishes that preterm infants, similar to term infants, generate a poliovirus type 3–specific T cell response and that the frequency of circulating poliovirus-specific responder T cells is comparable between preterm and term infants.
The study however, detected differences in circulating T cell populations between term and preterm infants.

Function of miR-146a in controlling Treg cell mediated regulation of Th1 responses

Function of miR-146a in controlling Treg cell mediated regulation of Th1 responses
I recently had the pleasure of listening an interesting talk on the function of micro RNAs in controlling the function of Tregs. The presenter was Dr. Li-Fan Lu from Howard Hughes Medical Institute and Immunology Program, Memorial Sloan-Kettering Cancer Center, New York, NY. Presented here is a brief blog on one of his paper.
Regulatory T cells (Tregs) have a suppressive function in body. They limit “collateral damage” resulting from protective immunity to infection and suppress sterile inflammation as well as unwanted immune responses to “self” and allergens. Tregs express FoxP3, an X chromosome encoded transcription factor which plays an important role in differentiation, homeostasis and function of Treg cells. It has been observed that FoxP3 directly or indirectly controls several thousand genes in Tregs. These genes are called as FoxP3 amplified genes many genes that serve as hallmark of Tregs and play important roles in their homeostasis and function. Also included among the FoxP3 amplified genes are those that code for noncoding RNAs called micro RNA (miRNA). These miRNAs serve as important regulators of Treg homeostasis and function both in basal and inflammatory settings. One such miRNA is miRNA-155, the cancer associated miRNA, which is constitutively expressed in high amounts in Treg cells in a FoxP3 dependent manner. It is transiently upregulated in T cells lacking FoxP3, B cells and myeloid cells up on activation. The investigators of this paper have previously shown that miR-155 causes increased response in Tregs to their key survival and growth molecule, IL-2 and is thus responsible for maintaining the numbers of Tregs. Another miRNA constitutively expressed in high amounts in Tregs is miR-146a. Like miR-155, it is induced up on activation of effector T cells and myeloid cells. In the present paper the authors tend to determine the role of miR-146a in Tregs. Their results are as follows:
• Elevated miR-146a expression in Tregs:
miRNA expressions were compared among FoxP3+ Treg cells and FoxP3- non Treg CD4+ cells---miR-146a was more expressed in Tregs in comparison to non-Tregs.
• miR-146a deficiency resulted in increased numbers but impaired function of Tregs:
Firstly, the authors examined FoxP3+ Treg cell subsets in the thymus and in the peripheral lymphoid organs of miR-146a deficient mice. They observed comparable sizes of different thymocyte and peripheral lymphoid and myeloid cell subsets and similar T cell activation status among the wild type and mutant mice. The number of FoxP3+ Tregs in mutant mice was higher in the periphery (but not in thymus) in comparison to wild type mice. These FoxP3+ Tregs also showed heightened proliferative activity as shown by increased expression of Ki67 and increase in other activation markers.
In next set of experiments, the authors transferred mixed bone marrow (1:1ratio of miR 146a deficient or sufficient Tregs mixed with Ly5.1 B6 or Ly5.1 FOXP3 KO) into Rag2-/- (or TCRβδ -/-) mice. They observed that in the mice group Mirn 146a-/-/FOXP3 KO, 40% of the mice died before the day of analysis. In these mice, all Tregs lack mir 146a as FOXP3 KO BM cells fail to generate Tregs; all other BM derived cells originate from both miR 146a deficient and sufficient BMs. The purpose of these experiments was to generate a miR 146a sufficient environment and then study the function of miR146a deficient Tregs in that environment. These mice showed severe clinical conditions and only 40% survived. Histologicaly also, these mice showed massive lymphocyte activation and tissue infiltraion. In contrast, mir 146a-/-/B6 mice had both miR 146a deficient and sufficient Tregs and they showed no clinical or histological signs of immune mediated lesions. So what was the cause of the adverse clinical and histological conditions in the mir 146a-/-/FOXP3 KO mice? The authors observed that the cause of these effects was not due to decrease in number of Tregs as the number of Tregs in KO mice was actually more than in control mice (mir 146a +/+/FOXP3 KO). The mir 146a -/-/ B6 mouse also showed an increase in miR 146a deficient Tregs than in controls (mir 146+/+/B6). These results suggested that miR146a deficiency increases the number of Tregs.
The authors next studied the activation of miR 146a sufficient T effector cells and observed a highly proliferative activity (Ki67), an increased CD62L low cell subset and high levels of activation markers ICOS and CTLA4 in FOXP 3 KO mice. However, both miR 146a deficient or sufficient cells showed control activation in a miR 146 a sufficient environment (B6 Ly5.1+ MICE). These results suggest that miR146 a play an important role in Treg mediated immunological tolerance.

• miR-146a deficiency in Treg cells resulted in dysregulated IFNγ responses: In the next set of experiments the authors isolated miR-146a deficient or sufficient Tregs from mir 146a-/-/B6 or mir146a-/-/FOXP3 KO mice and co-cultured them with wild type responder CD4+ T cells (in the presence of CD3 antibody and irradiated T cell depleted splenocytes). They observed that miR 146a deficient Tregs from mir 146a-/-/B6 mice were as suppressive as miR 146a sufficient Tregs from control animals (mir146a+/+/B6). However, miR 146a deficient Tregs from knock-outs were more suppressive than miR146a sufficient Tregs from controls. These results indicate that deficiency of miR 146a did not diminish the overall suppressive activity of Tregs. They also observed that loss of miR146a in Tregs resulted in increase production of pro-inflammatory cytokine IFNγ by both miR 146a sufficient and deficient Cd4 and CD8 cells. This increase was not observed in the presence of miR146a sufficient Tregs.

• IFNγ blockade prevents the autoimmune disease in mice harboring miR-146a deficient Tregs: The next question was weather this increased production of IFNγ is responsible for clinical condition in Mirn 146a-/-/FOXP3 KO mice. To answer this question, the investigators performed experiments in which IFNγ was blocked by treating these mice with IFNγ blocking antibody. The increase in number of miR 146a deficient Tregs was not affected in the presence of blocking antibody. However, the blocking of IFNγ prevented disease in deficient mice. The blockade of IFNγ with blocking antibody did not cause simultaneous increase in IL4 and IL-17 (Th2 responses). These results suggest that the diseases observed in the miR146a deficient Tregs was IFN-gamma dependent and blockade of IFN-gamma can control Th2 and Th17 responses but not Th1 response.

• miR 146a regulates stat1 in Tregs: Stat 1 is a key transcription factor downstream of IFN-gamma signaling. Thus, the authors transfected CD4+ T cells isolated from miR146a sufficient or deficient with a luciferase reporter construct containing wild type or mutant Stat1 and assessed luciferase activity. miR146a sufficient T but not deficient cells transfected with wild type Stat1 showed suppression of reporter activity. The authors also performed immunoblot analysis of Stat1 protein in Tregs and non-Tregs. There was an increase in production of Stat1 in Mir146a deficient Tregs and non-Tregs.

• miR 146a controls Treg cell mediated regulation of IFNγ responses through targeting Stat1: Based on previous experiments the authors hypothesized that miR146a deficiency causes dysregulated IFN-gamma production, which in turn causes increased expression of Stat1, which can then lead to failure of Tregs to control Th1 response. To test this hypothesis, the authors generated miR146a deficient mice harboring a single function stat1 allele (Mirn 146a-/-/Stat1+/-). The BM cells from these mice were mixed with Ly5.1+ FOXP3 KO mice and transferred into Rag2-/- mice. Thus, the recipients are Mirn 146a-/-/Stat1+/-/FOXP3 KO. Then they measured the amount of Stat1 phosphorylated in Tregs and non-Tregs from these and from control mice (Mirn 146a-/-/Stat1+/+/FOXP3 KO). Both Tregs and non-Tregs from test mice showed nearly 50% reduction in phosphorylated stat1 in comparison to controls. The test mice also developed a much milder and delayed disease. The authors also observed significant reduction in production of IFN-gamma in tests mice

Source: Lu LF, Boldin MP, Chaudhry A, Lin LL, Taganov KD, Hanada T, Yoshimura A, Baltimore D, Rudensky AY. Function of miR-146a in controlling Treg cell-mediated regulation of Th1 responses. Cell. 2010 Sep 17;142(6):914-29. PubMed PMID: 20850013; PubMed Central PMCID: PMC3049116.