Longitudinal tracking of TREE-directed transgene expression in injured mice by IVIS

(A) Schematic illustration of study design. Albino BALB/c mice were systemically injected with AAV9 vectors packaging fLuc reporter cassettes directed by TREEs and a permissive promoter. Mice underwent I/R surgery at D61 and were imaged by IVIS at indicated time points. (B) Representative IVIS images indicate changes of expression over time and space for each vector. Cardiac region of interest (ROI) indicated by red box. n = 2 mice. (C) Average radiance measured from cardiac ROIs plotted over days post-injury (dpi). Average radiance normalized to their baseline pre-injury were also plotted (right). n = 2 mice. (D) Average cardiac radiance showed a transient increase in expression for both REN (left) and 2ankrd1aEN (right) after I/R injury, whereas sham operated animals showed relatively constant expression. n = 2 mice for I/R, n = 3 mice for sham.

Liver-de-targeted AAV.cc84 capsid limits hepatic expression from TREEs

(A) Schematic of experimental timeline, comparing AAV9 and AAV.cc84 capsids for systemic delivery of REN-Hsp68::fLuc. Mice were IVIS imaged in the weeks following AAV delivery and post-sham surgery. (B) Representative IVIS images of mice injected with either AAV9 (left) or AAV.cc84 (right) at 14 (top) and 21 days (bottom) post-AAV injection. Mice were also subdivided by biological sex to account for sex differences in AAV liver tropism. (C) Average radiance from liver ROIs showed significantly higher expression in AAV9-transduced mice compared to AAV.cc84 through all timepoints (n = 6 mice, Holm-Sidak multiple comparisons test). (D) Representative IVIS images of harvested organs at 42 days post-AAV demonstrate liver expression with AAV9 (left) while undetected with AAV.cc84 (right). (E) Vector genome quantification from collected liver samples reveal higher liver transduction with AAV9 compared to AAV.cc84 for both female and male mice.

Post-injury delivery of AAV.cc84-packaged 2ankrd1aEN targets expression to cardiac injuries

(A) Schematic of experimental timeline comparing expression between Hsp68 and 2ankrd1aEN when delivered at 4 dpi. (B) Representative IVIS images of mice injected with Hsp68 (top) or 2ankrd1aEN-Hsp68 (bottom) after I/R injury. (C) Cardiac average radiance normalized to the 7 dpi time point increased over time with 2ankrd1aEN while remaining stable with Hsp68 (n = 4 mice, Holm-Sidak multiple comparisons test). (D) Average cardiac radiance directed by 2ankrd1aEN was significantly higher in mice with I/R injury compared to Sham at 7 dpi (n = 4 mice, Welch’s t test). (E) Average cardiac radiance was more significantly elevated in the first 21 days post-injury in mice with I/R injury compared to sham (n = 4 mice, Mann-Whitney tests).

Screening of AAV libraries for enriched capsids in injured myocardium

(A) Schematic of AAV capsid library screening delivered systemically to mice with either sham (n = 1 mouse) or I/R (n = 2 mice) injury at 9 dpi. Two days later, hearts were biopsied to collect AAV genomes in either injured or remote tissues. (B) Capsid sequenced reads enriched in the injured tissues were plotted against sham animals. Each point represents a unique capsid. Wild-type AAV9 capsid is marked by blue arrow. (C) Representative fluorescence imaging of AAV9 (top) or variant capsid IR41 (bottom) delivering a self-complementary CBA::EGFP cassette at 16 dpi. Asterisks indicate infarct site, imaged at higher magnification in middle and right panel. Dashed white lines indicate the border zone region. Left scale bar, 1 mm. Middle and right scale bar, 100 um.

Post-injury delivery of AAV.IR41 variant capsid enhances 2ankrd1aEN-directed expression in injured myocardium

(A) Representative IVIS images of mice with sham (top) or (bottom) surgery transduced with either AAV9 (left) or IR41 (right) packaged with 2ankrd1aEN-Hsp68::fLuc (n = 3-4 mice). (B, C) Cardiac average (B) and maximum (C) radiance was elevated in MI mice transduced with IR41 compared to AAV9 (n = 3-4 mice, Holm-Sidak multiple comparisons test). (D) Viral genome quantification from heart tissues was elevated in MI mice injected with IR41 compared to AAV9 (n = 3-4 mice, Holm-Sidak multiple comparisons test).

IVIS imaging for tracking spatiotemporal expression of rAAV vectors

(A) Schematic showing comparison of AAV9 vectors packaging either the strong, constitutively active chicken beta actin (CBA) promoter or minimal heat shock protein 68 (Hsp68) promoter to direct fLuc expression. n = 5 mice/AAV group. (B) Representative IVIS images of mice injected with AAV containing either CBA (top) or Hsp68 (bottom) promoters. Red boxes indicate ROIs marking cardiac, liver, and whole-body expression. (C) Average radiance measured from cardiac ROIs from CBA (top) or Hsp68 (bottom) promoters show relatively consistent expression from 30-68 days post-AAV injection. Square, male mice. Circle, female mice. (D, E) Average radiance measured from liver (D) and whole body (E) ROIs showed relatively consistent levels of expression over time for both promoters.

AAV.cc84 capsid retains cardiac tropism while minimizing liver transduction

(A) Representative IVIS images of mice injected with either AAV9 (left) or AAV.cc84 (right) at 7 (top) and 35 days (bottom) post-AAV injection. (B) Average radiance in the heart was similar between AAV9 and AAV.cc84 (n = 6 mice, Holm-Sidak multiple comparisons test). (C) Vector genome quantification from cardiac tissues showed similar levels of vector genomes between AAV9 and AAV.cc84 for both sexes (n = 3 mice, Holm-Sidak multiple comparisons test). (D) Representative IVIS images of mice with ROIs used to measure expression in head/neck and lower abdomen (white dashed box). (E) Average radiance measured in the head/neck (left) and lower abdomen (right) was similar between AAV9 and AAV.cc84 over the course of the study (n = 6 mice, Holm-Sidak multiple comparisons test).

Delivery of AAV.cc84 packaging 2ankrd1aEN after myocardial injury

(A) I/R surgery injury extent was assessed by ejection fraction via echocardiography to estimate injury prior to AAV delivery (n = 4 mice). (B) Cardiac average radiance plotted for each individual mouse with I/R injury plotted over time with either Hsp68::fLuc (gray) or 2ankrd1aEN-Hsp68::fLuc (black). (C) Representative IVIS images of sham-operated mice injected with AAV.cc84 packaged with 2ankrd1aEN-Hsp68::fLuc.

Screening AAV capsid libraries enriched in injured myocardium

(A) Injury site-enriched capsids in I/R-injured mice. Each point represents a unique capsid. Top left quadrant, capsids co-enriched at injury site. Wild-type AAV9 capsid is marked by an orange arrow. (B) Representative fluorescence imaging of variant capsids, IR42 (left) and IR43 (right), delivering a self-complementary CBA::EGFP cassette at 16 dpi. Scale bar, 1 mm. (C) Immunofluorescence of infarct sites of mice transduced with either AAV9 (left) or IR41 (right) show no colocalization of EGFP and Vimentin. Scale bar, 100 um.

Post-injury delivery of variantAAV.IR41 variant capsid enhances 2ankrd1aEN-directed expression in injured myocardium over AAV9

(A) Compiled IVIS images of mice that underwent sham or MI surgery with AAV9 or IR41 transduced at 3 dpi. Mice were imaged at 7, 14, and 21 dpi. (B, C) Cardiac average (B) and maximum (C) radiance were statistically similar between AAV9 and IR41 in sham-operated mice.