These experiments demonstrate the requirement for human BLyS to efficiently reconstitute human B cells in NOD rag1−/−Ppf−/−
immunodeficient mice. A species specific BLyS restriction is demonstrated for human B cells in culture and for B cells transferred to murine hosts. The NOD rag1−/−Ppf−/−
environment is not BLyS deficient, per se, as these mice readily support the engraftment of murine B cells. Species specificity in the action of survival and growth promoting cytokines is not novel to members of the TNF family; IL-2, IL-5, IL-6, and gamma interferon have all demonstrated species restrictions 
. Moreover, amino acid differences between murine and human BLyS in the BR3 receptor binding region and/or differences in the BLyS binding region of the BR3 receptor are known to have marked effects on survival signaling for human B cells 
. While we favor the notion that the basis for species restriction is in the interaction between BLyS and BR3, human B and T cells also express the TACI receptor for BLyS and differences in huBLyS and muBLyS binding to and signaling through TACI may also contribute to our results.
In vitro, human BLyS supported significantly better human B cell survival than did murine BLyS, which was indistinguishable from unsupplemented cultures. In contrast, human and murine BLyS were equally effective at promoting the in vitro survival of murine B cells. Concurrent with the survival effect, BLyS stimulation leads to a prolonged induction of NF-κB p100 processing and nuclear localization of NF-κB p52 (laboratory observations and 
). While freshly isolated human B cells bind FLAG-tagged murine BLyS as efficiently as human BLyS, p52 nuclear localization in human B cells induced by human BLyS was more robust than that induced by murine BLyS. This shows a dissociation between BLyS binding and survival signaling which may be the result of a species difference in the binding requirement for activation of BR3/TACI receptors on human B cells. It is noteworthy that in titration of human BLyS on murine B cells diminished B cell survival was correlated directly with the extent of p52 nuclear localization (supplemental ). Our laboratory has recently defined the survival pathways induced by BLyS in murine B cells 
and further experiments are in progress to define these pathways in human B cells.
Engraftment of human B cells is efficient in huBLyS treated NOD rag1−/−Ppf−/−
recipients. Human BLyS is continually required to maintain the B cells over the 3 week period used in these studies since treatment with a BLyS decoy receptor, TACI-Ig, which depletes both human and murine BLyS from the animals, resulted in loss of B cells as assessed by immunohistological analysis and resulted in lower serum IgM and IgG compared with mice treated for 14 days with BLyS. Our engraftment data using huBLyS supplementation contrasts from that observed in other PBL xenochimera models where few, if any, human B cells are observed after cell transfer and those B cells do not survive beyond one week 
The B cells are functional as we find significantly higher amounts of human IgM and IgG in chimeras receiving huBLyS compared with PBS treated controls. Human Ig production could result from the homeostatic activation of naïve or memory B cells. Homeostatic B cell proliferation is seen when murine B cells are transferred into B cell deficient environments 
, however this mechanism has not been evaluate in xenochimeras. It is also possible that activation by xenoantigens can contribute to B cell activation and serum Ig. The antigen responsiveness of engrafted human B cells and the ability to induce a de novo immune response is demonstrated by the production of anti-pneumococcal antibodies following immunization with the thymus-independent antigens in pneumovax23. There was a 10–20 fold increase, respectively in IgM and IgG, in pneumococcal antibodies in BLyS treated recipients compared to BLyS treated and unimmunized and PBS treated immunized controls. This xenochimera model with enhanced B cell engraftment provides a unique opportunity to test vaccine responses using PBL samples from a variety of individuals, including neonates and the aged who frequently exhibit weak immune responsiveness.
Immunohistological analysis of recipient mice receiving only PBS revealed only scattered human T cells but these cells were readily identified by FACS analysis of collagenase digested spleens. It was noted, however, that there was a marked improvement of T cell engraftment with human BLyS supplementation. Both T and B cell yields from BLyS treated recipients was much higher than that observed with PBS treated recipients. Recent publications demonstrate that BLyS can act as a co-stimulator for T cells acting through BR3 and/or TACI 
. T cells, like B cells, will under go homeostatic proliferation when introduced to a T cell deficient environment 
. This proliferation causes induction of early activation markers, including expression of BR3 mRNA and protein on T cells 
. Human T cells may also undergo activation to xenoantigens in the murine environment, again upregulating the BLyS receptor. In this regard, it is also possible that the human B cells themselves can present antigen to the T cells, thereby facilitating T cell expansion and survival. TACI-BLyS signaling in B cell-dendritic cell interactions have been shown to increase the expansion of antigen responsive CD8+
T cells 
whereas BR3-BLyS costimulates CD4+
T cell alloresponses 
. We did note that use of the TACI decoy receptor in vivo resulted in loss of both B and T cells, although B cells were affected to a greater degree. Taken together, these data suggest that human T cell engraftment may also be enhanced directly or indirectly by human BLyS treatment. Further studies using purified human T cells and human BLyS supplementation should address whether human BLyS acts directly on T cells to facilitate engraftment.
Since huBLyS enhances engraftment of both B and T cells, we considered the possibility that prolonged exposure to huBLyS may increase the rate of graft verses host disease (GVH) in the PBL recipient mice. No GVH was seen by observation or histological analysis at the 3 week time points assessed in these experiments even in recipients with the robust B and T cell engraftment, however, other investigators have found evidence of GVH by 4–6 weeks post-PBL engraftment in a variety of immunodeficient murine hosts 
. Regulatory T cells have recently been shown to express BLyS receptors and it is possible that huBLyS treated animals engraft sufficient regulatory cells to delay or prevent GVH 
Our data suggest recombinant human BLyS has a significant enhancing effect on the engraftment of both T and B cell populations in PBL.