Supporting Information for E—Y Kim et al.

Supplementary information for

Gold Nanoparticle—Mediated Gene Delivery Induces Widespread Changes in the Expression of Innate Immunity Genes

Eun—Young Kim1, Reiner Schulz2, Peter Swantek1, Kevin Kunstman1, Michael H. Malim3, and Steven M. Wolinsky1*

*To whom correspondence should be addressed. E—mail: s—

Supporting figures and tables

Figure S1. The UV-VISabsorption spectrum of the gold nanoparticles stably functionalized with covalently attached oligonucleotides. We observed no shifts in the plasmon resonance of the gold nanoparticle oligonucleotide complexes under the experimental conditions we used. Shown is a spectrum for a representative gold nanoparticle oligonucleotide complex preparation.

Figure S2. Transmission electron microscope image of the gold nanoparticles stably functionalized with covalently attached oligonucleotides.The gold nanoparticle oligonucleotide complexes were distributed as discrete, highly electron—dense particles. There was no change in the gold nanoparticle oligonucleotide complex preparations over time. Thus, the gold nanoparticles stably functionalized with covalently attached oligonucleotides remained stable.

Figure S3. The numbers of probe sets detecting significant (P—value < 0.01) differentialexpression in PBMCs treated with gold nanoparticle oligonucleotidecomplexes that exceeds a given fold change threshold. The log2 foldchange threshold is plotted along the x—axis and the number of probesets that exceed the threshold in terms of either down—regulation (left)or up—regulation (right), respectively, are plotted on the y—axis (Table S2).

Figure S4. The most overrepresented KEGG pathways in PBMCs treated with gold nanoparticle oligonucleotide complexes.All differentially expressed genes (SAM multiple testing corrected P—value < 0.01) changed gene expression by at least 20% (up—regulated [pink], up—regulated at least two fold [red], down—regulated [cyan] and down—regulated by at least 50% [blue]). A box in a KEGG pathway diagram can correspond to a single gene/protein, a multi-gene protein complex, or a group of genes/proteins/protein complexes, and therefore it can have multiple colors to indicate both up— and down—regulation: a, lysosome; b, cytokine-cytokine receptor interaction; c, hemotopoetic cell lineage; d, NOD—like receptor signaling; and e, chemokine signaling.

Figure S5. The most overrepresented GO categories in293T cells treated with gold nanoparticle oligonucleotide complexes. Shown are the –log10—transformed raw P—values of the 10 most overrepresented GO categories for two significance thresholds of differential gene expression: multiple testing corrected SAM P—value < 0.05 (blue) versus SAM P—value < 0.0125 (red). There was no overlap of the most overrepresented GO categories or KEGG pathways between PBMCsand 293T cells.

Table S1. Microarray conditions. Each number or pair of numbers in parentheses is a microarray where the first number denotes the microarray batch, and the second number denotes thegold nanoparticle oligonucleotide complex batch if applicable. The common subset of treatment conditions among the microarray experiments specifically control for non—biological experimental variation due to various causes. The evaluation for batch effects using principal component analysis indicate that the differences in gene expression between PBMCs treated with gold nanoparticle oligonucleotide complexes and untreated or otherwise treated PBMCs does not correlate with batch effects or other systematic bias in the microarray expression data. See file: Table_S1.doc.

Table S2. SAM differential gene expression analysis results for PBMCs treated with gold nanoparticles stably functionalized with covalently attached oligonucleotides. Only informative probe sets (I/NI P—value < 0.6) were included in the analysis. The first tab, named 'all', contains all data for all probe sets, ranked by significance of the observed differential expression in PBMCs (no differential expression data for non-informative probe sets). For the probe sets that detected significant (P—value 0.01) differential expression, the two tabscontain the data for only those probe sets that were either up— (n = 401) or down—regulated (n = 529) at least two—fold ('>2fold up' and '>2fold dn',respectively) ranked by their fold—change. See file: Table_S2_SOM_AuNP_PBMC_I_log2_SAM.xls.

Table S3. Final Fold Enrichment (Relative Quantity 2-ΔΔCt) compared with controls. The fold changes of gene expressions in 293T cells were not significant between the gold nanoparticle oligonucleotide complextreated and untreated groups (6—hr, median 1.45, range 1.21—1.85; 24—hr, median 1.32, range 0.84—2.57; and 48—hr, median 1.285, range 0.85—2.0). Gene expression in primary immune cells in response to the gold nanoparticle oligonucleotide complexes showed greater fold changes (6—hr, median 6.58 fold change, range 1.26—53.92; 24—hr, median 2.86 fold change, range 0.15—160.8; and 48—hr, median 3.38 fold change, range 0.34—223.3) than the other conditions. The principal changes were in genes of the TLR signaling, T cell activation, endocytosis, and apoptosis pathways. Among these genes, IL—1b was up—regulated the most (223.3 fold change) in PBMCs in response to the gold nanoparticle oligonucleotide complexes. CTLA4, IL1B, IL2RA, and TLR4 expression was not detected (ND) in 293T cells, highlighting the cell type specific differences between PBMCs and lineage—specific cell types. The RT—qPCR and the human microarray gene expression results were equivalent. The 12 selected genes were: clathrin, light chain (CLTA); CTLA4 cytotoxic T-lymphocyte—associated protein 4 (CTLA4); EH—domain containing 1 (EHD1); FYN oncogene related to SRC, FGR,YES (FYN); interleukin 1, beta (IL1b); IL2RA interleukin 2 receptor, alpha; jun oncogene (JUN); mitogen-activated protein kinase kinase 7 (MAP2K7); nuclear factor of kappa light polypeptide gene enhancer in B-cells 2 (p49/p100) (NFKBIA); nuclear factor of kappa light polypeptide gene enhancer in B—cells inhibitor, alpha (NFKB2); toll—like receptor 1 (TLR1); and toll—like receptor 4 (TLR4). See file: Table_S3.doc.

Table S4. NCBI DAVID GO category overrepresentation analysis results for PBMCs treated with gold nanoparticles stably functionalized with covalently attached oligonucleotides.See file: Table_S4_SOM_AuNP_PBMC_SAM_BH_GO.xls.

Table S5. NCBI DAVID KEGG pathway overrepresentation analysis results for PBMCs treated with gold nanoparticles stably functionalized with covalently attached oligonucleotides.See file:Table_S5_SOM_AuNP_PBMC_SAM_BH_KEGG.xls.

Table S6. SAM differential gene expression analysis results for 293T cellstreated with gold nanoparticles stably functionalized with covalently attached oligonucleotides.(negative control and 6—hours versus 24— and 48—hours). Only informative probe sets (I/NI P—value < 0.6) were included in the analysis. See file: Table_S6_SOM_AuNP_293T_I_log2_SAM.xls.

Table S7. NCBI DAVID GO category overrepresentation analysis results for 293T cells treated with gold nanoparticles stably functionalized with covalently attached oligonucleotides.See file: Table_S7_SOM_AuNP_293T_SAM_BH_GO.xls.

Table S8. NCBI DAVID KEGG pathway overrepresentation analysis results for 293T cells treated with gold nanoparticles stably functionalized with covalently attached oligonucleotides.See file: Table_S8_SOM_AuNP_293T_SAM_BH_KEGG.xls.