Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Molecular Sciences

Research Advisor

Malak Y. Kotb, Ph.D.


David Brand, Ph.D. Marko Radic, Ph.D. Dario Vignali, Ph.D. Ae-Kyung Yi, Ph.D.


Group A streptococci, Superantigens, HLA class II polymorphism, TCR Vb repertoire variation, Streptococcal toxic shock syndrome, transgenic mice


Group A streptococci (GAS) are serious human pathogens that can cause a wide array of diseases ranging from pharyngitis to streptococcal toxic shock (STSS), which is caused by excessive cytokines responses triggered by streptococcal superantigens (Strep SAgs). SAgs interact simultaneously with HLA class II molecules on antigen presenting cells (APC) and with the T cell receptor beta chain variable elements (TCR V) and cause excessive stimulation of both cell types and massive release of inflammatory mediators.

The Strep SAgs include the streptococcal pyrogenic exotoxins (Spes), which play a crucial role in the pathogenesis of severe invasive streptococcal diseases, including Strep toxic shock syndrome (STSS). However, not all septic individuals infected with SAg producing GAS develop STSS. Our epidemiologic studies showed that individual HLA class II allelic variation play a crucial role in modulating the severity of invasive GAS diseases. Certain HLA class II alleles and haplotypes confer strong protection from STSS (DRB1*1501/DQB1*0602) or increase the risk for it (DRB1*1401/DQB1*0503 and DRB1*0701/DQB1*0202), whereas others behave in a neutral fashion with respect to disease severity. The main aim of this thesis was to validate the direct role of HLA class II variation in modulating the severity of GAS sepsis and investigate molecular mechanisms underlying the differential effect of HLA class II variation on SAgs responses in vitro and in vivo.

We validated our epidemiological findings and confirmed the direct role of HLA class II allelic variation in modulating the severity of GAS sepsis using HLA transgenic (tg) mice expressing alleles of interest. Splenocytes from mice expressing the protective HLA-DQB1*06 (DQ6) allele mounted significantly lower proliferative and cytokine responses to Strep SAgs, than splenocytes from mice expressing the neutral DRB1*0402/DQB1*0302 (DR4/DQ8) alleles. Furthermore, cross presentation of Strep SAgs to pure T cells by antigen presenting cells (APC) from these HLA-tg mice showed that presentation by HLA-DQ6 APC elicited significantly lower proliferative and inflammatory cytokine responses than by the HLA-DR4/DQ8 APC. These in vitro data were supported by in vivo findings, as the DQ6 mice showed significant resistance to GAS sepsis with lower inflammatory cytokine responses and longer survival, than the DR4/DQ8 mice.

Similarly, presentation of the Strep SAgs by human APC expressing STSS protective haplotypes elicited significantly lower proliferative and cytokine responses than by APC expressing neutral or high risk haplotypes, even when the responding APC-depleted T cells came from different individuals with different TCR V repertoires. Moreover, we screened SAgs responses over a panel of B lymphoblastiod cell lines (BLCL), expressing the most prevalent HLA class II haplotypes in populations with high incidence of STSS, using pure T cells from different individuals, and reproducibly observed significantly higher responses in the presence of the high-risk haplotypes or neutral than protective haplotypes. These data demonstrate a dominant role of individual HLA class II and not TCR Vß repertoire variation in determining the potency or type of responses to the Strep SAgs.

An important finding from the above studies is that the presentation of the Strep SAgs by the protective HLA class II haplotype (DR15/DQ6) elicits high levels of the regulatory cytokine IL-10, and thus may be actively suppressing inflammatory cytokine production. To further investigate this point and to determine if in the clinical setting, humans heterozygous for high risk and protective haplotypes would be protected or prone to STSS, we generated a panel of bare lymphocyte syndrome cells (BLS) (class II negative cells) expressing individual HLA-DR, DQ to delineate their relative contribution to the potency and type of Strep SAg responses. In addition to the mixture of M1T1 SAgs, we also generated a panel of recombinant SAgs found in this mixture to investigate the interaction and relative contribution of variable HLA and SAg molecules to the overall response.

To ensure equimolar expression of HLA class II alleles on BLS cells, we utilized the novel technology of ‘self cleaving’ 2A peptide to generate multicistronic retroviral vectors. The constructs were cloned into pHSPG retroviral vector and packaged into HSPG viral particles that were used to transduce (BLS). DR or DQ and chains were expressed as a single message linked by 2A peptide resulting in coordinated expression of both chains that was confirmed by quantitative flow cytometric analysis of the transduced BLS cells.

Presentation of individual Strep SAgs by BLS cells expressing either of the protective alleles (DR15 or DQ6) induced significantly lower proliferative and cytokine responses than that by BLS cells expressing single alleles on the high-risk haplotypes. Reproducibly, higher levels of the immunosuppressive cytokine IL-10 were produced in the presence of the protective alleles. Thus, confirming our suggestion that protective alleles mediate their effects through active suppression of inflammatory cytokines production.

In an attempt to understand the molecular basis, we investigated if differences in binding affinities between specific SAgs and HLA class II alleles affect response level. In general, there was a correlation between binding affinity of Alexa fluor 647-labeled SAgs to individual class II alleles and response levels, except for SmeZ1 binding to and presentation by DQ alleles, where an inverse correlation was observed. However, Ca flux studies demonstrated good correlation with the observed responses. These data indicate that individual class II alleles differentially bind and present Strep SAgs resulting in significantly different proliferative and cytokines responses. The inverse relation between SmeZ1 responses and its binding affinities to DQ alleles suggest that different combinations of SAgs-HLA alleles may trigger distinct signaling in T cells resulting in variable levels of response.

In conclusion, our study demonstrate dominant and direct role of HLA class II in controlling Strep SAgs responses and resulting disease severity. The data indicate that each SAg-class allele is unique with distinct behavior that cannot be extrapolated to others. Future studies will investigate this point further and determine the interactions between high risk and protective HLA class II alleles, when expressed on the same APC.