Development of a PCSK9-targeted nanoparticle vaccine to effectively decrease the hypercholesterolemia

Abstract

Proprotein convertase subtilisin/kexin type 9 (PCSK9) binds to the low-density lipoprotein receptor (LDLR) and mediates its internalization and degradation, resulting in an increase in LDL cholesterol levels. Recently, PCSK9 emerged as a therapeutic target for hypercholesterolemia and atherosclerosis. In this study, we develop a PCSK9 nanoparticle (NP) vaccine by covalently conjugating the catalytic domain (aa 153-aa 454, D374Y) of PCSK9 to self-assembled 24-mer ferritin NPs. We demonstrate that the PCSK9 NP vaccine effectively induces interfering antibodies against PCSK9 and reduces serum lipids levels in both a high-fat diet-induced hypercholesterolemia model and an adeno-associated virus-hPCSK9^D374Y-induced hypercholesterolemia model. Additionally, the vaccine significantly reduces plaque lesion areas in the aorta and macrophages infiltration in an atherosclerosis mouse model. Furthermore, we discover that the vaccine's efficacy relied on T follicular help cells and LDLR. Overall, these findings suggest that the PCSK9 NP vaccine holds promise as an effective treatment for hypercholesterolemia and atherosclerosis.

Keywords: LDLR; PCSK9; T follicular help cells; hypercholesterolemia; nanoparticle vaccine.

Graphical abstract

Figure 1

Characterization of the PCSK9 catalytic domain (D374Y) NP vaccine (A) Schematic representation of a NP vaccine based on a ferritin self-assembling system (HPF), predicted by AlphaFold, displaying the catalytic domain of PCSK9 with the D374Y mutation. (B) Coomassie blue staining of GvO-HPF, Sd-PCSK9, and PCSK9-NP, showing distinct protein bands. (C) Representative elution chromatogram of PCSK9-NP using a pre-packed Superose 6 column. Ultraviolet absorption at 280 nm was observed. (D) Transmission electron micrograph of PCSK9-NP displaying the morphology of the NPs. GvO, GvTagOpti; mAU, milli-absorbance units; PCSK9-NP, PCSK9 NP; Sd, SdCatcher.

Figure 2

Induction of potent immune responses by the PCSK9-NP vaccine in BALB/c mice and beagles (A) Schematic representation of the BALB/c mice vaccination protocol. Mice received two subcutaneous immunizations at W0 and W4. Serum samples were collected biweekly. (B) Time-course curve of PCSK9 catalytic domain-specific IgG titers measured in mouse sera. (C) Calculation of endpoint titers of anti-PCSK9 catalytic domain IgG1, IgG2a, and IgG2b at W6 using ELISA. (D) Detection of IgG antibodies against full-length PCSK9 at W6. (E) Flow cytometry (left) and statistical analyses (right) showing the percentages and absolute numbers of CD4⁺CXCR5⁺PD1⁺ Tfh cells (top) and CD19⁺Fas⁺GL7⁺ GCB cells (bottom) in the draining lymph nodes at W2. (F) Schematic representation of the beagle vaccination protocol. (G) Time-course curve displaying PCSK9 catalytic domain-specific IgG titers in sera from beagles. (H) Flowchart illustrating the cell-based experiments conducted to evaluate the in vitro interfering activity of serum antibodies. (I and J) Flow cytometry (left) and statistical analyses (right) showing the MFI of the cell-surface receptor LDLR in Huh7 cells exposed to PCSK9 in conjunction with sera from immunized BALB/c mice (I) and from beagles at W6 (J). Experiments were conducted independently in triplicates; all data are presented as mean ± SEM; n = 4–5 per group. Data in (C–E) and (I) were analyzed by one-way ANOVA with Tukey’s multiple comparisons test. two-way ANOVA with Tukey’s (B) and Sidak’s (G) multiple comparisons test were used for statistical analysis. An unpaired Student’s t test (J) was used. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. ns, no significance. See also Figures S1, S2, S4, and S5.

Figure 3

The PCSK9-NP vaccine prevents high-fat diet-induced hypercholesterolemia in C57BL/6-hPCSK9 mice (A) Schematic representation of the experiment procedure. C57BL/6-hPCSK9 mice were subcutaneously immunized twice, followed by initiation of a high-fat diet from W4 onward. (B) Measurement of PCSK9 catalytic domain-specific IgG titers. (C) Measurement of T-CHO (left), LDL-C (middle left), TG (middle right), and HDL-C (right) levels in mouse sera at various time points using a detection kit. (D) Representative (left) and statistical analyses (right) H&E (top) and Oil Red O staining (bottom) of liver tissues. (E) Measurement of T-CHO (top) and TG (bottom) levels in mouse liver tissues. (F) Immunofluorescent (IF) staining for LDLR on liver sections. Experiments were conducted independently in triplicates; all data are presented as mean ± SEM; n = 3–4 per group. Data in (B) and (C) were analyzed by two-way ANOVA with Sidak’s multiple comparisons test. An unpaired Student’s t test was used for statistical analyses in (D) and (E). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. See also Figure S3.

Figure 4

The PCSK9-NP vaccine prevents AAV-hPCSK9^(D374Y)-induced hypercholesterolemia in C57BL/6 mice (A) Schematic representation of the experiment procedure. C57BL/6 mice were subcutaneously immunized twice. At W4, the immunized mice received tail vein injections of 2 × 10¹⁰ vg of AAV-hPCSK9^(D374Y). (B) Measurement of PCSK9 catalytic domain-specific IgG titers. (C) Measurement of T-CHO (left), LDL-C (middle left), TG (middle right), and HDL-C (right) levels in mouse sera. (D) Representative (left) and statistical analyses (right) of Oil Red O staining of liver tissues. (E) Measurement of T-CHO (top) and TG (bottom) levels in mouse liver tissues. (Experiments were conducted independently in triplicates; all data are presented as mean ± SEM; n = 5 per group.) Two-way ANOVA with Sidak’s (B) and Tukey’s (C) multiple comparisons test were used for statistical analysis. An unpaired Student’s t test was used for statistical analyses in (D) and (E). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Figure 5

The reducing effect of the PCSK9-NP vaccine on hypercholesterolemia in mice depends on Tfh cells and the LDL receptor (A) Schematic representation of the experiment procedure. (B) Flow cytometry (left) and statistical analyses (right) showing the percentages and numbers of Tfh cells (top), GCB cells (middle), and B220⁻CD138⁺ plasma cells (bottom) in the draining lymph nodes at W2. (C) Measurement of PCSK9 catalytic domain-specific IgG titers. (D) Calculation of endpoint titers of anti-PCSK9 catalytic domain IgG1 (left), IgG2a (middle), and IgG2b (right). (E) Measurement of T-CHO in mouse sera. (F) Schematic representation of the experiment procedure. LDL receptor (LDLR) wild-type (LDLR^+/+) on a high-fat diet and LDLR knockout (LDLR^−/−) mice were subcutaneously immunized twice. (G) Measurement of PCSK9 catalytic domain-specific IgG titers in serum samples form LDLR^−/− mice. (H) Measurement of T-CHO in mouse sera. (Experiments were conducted independently in triplicates; all data are presented as mean ± SEM; n = 5–6 per group.) Data in (B) and (D) were analyzed by one-way ANOVA with Tukey’s multiple comparisons test. two-way ANOVA with Tukey’s (C and E) and Sidak’s (G and H) multiple comparisons test were used for statistical analysis. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. ns, no significance.

Figure 6

The PCSK9-NP vaccine alleviates the progression of atherosclerosis in the AAV-hPCSK9^(D374Y)-induced mouse model (A) Schematic representation of the experiment procedure. C57BL/6 mice were subcutaneously immunized three times at W0, W4, and W8. At W4, the immunized mice received an injection of 1 × 10¹¹ vg of AAV-hPCSK9^(D374Y) through the tail vein and were started on a Paigen diet beginning at W4. (B) Measurement of LDL-C in mouse sera at W14. (C) Visualization and quantification of aortic atherosclerotic lesions using Oil Red O staining. (D) Representative tissue-section image (left) and summary of data (right) Oil Red O staining (plaque area, top), H&E stating (necrotic area, middle), and Masson’s trichrome stating (collagen area, bottom) of the aortic roots. (E) Evaluation and statistical analyses of immunofluorescent (IF) staining for F4/80 (top) and α-smooth muscle actin (middle). (Experiments were conducted independently in triplicates; all data are presented as mean ± SEM; n = 5–6 per group.) An unpaired Student’s t test was used for statistical analysis. ∗p < 0.05. ns, no significance.

Figure 7

The PCSK9-NP vaccine treats hypercholesterolemia in different mouse models (A) Schematic presentation of the experiment procedure in high-fat diet-induced hypercholesterolemia. Mice were fed a high-fat diet starting 2 weeks prior to the initial immunization. (B) Determination of PCSK9 catalytic domain-specific IgG titers. (C) Measurement of both T-CHO (left) and LDL-C (right) levels in mouse sera. (D) Schematic representation of the therapeutic experiment procedure in AAV-hPCSK9^(D374Y)-induced hypercholesterolemia. Mice were injected with 2 × 10¹⁰ vg AAV-hPCSK9^(D374Y) through the tail vein before one week prior to the initial immunization. (E) Determination of PCSK9 catalytic domain-specific IgG titers. (F) Measurement of both T-CHO (left) and LDL-C (right) levels in mouse sera. (G) The time-course curves of PCSK9 catalytic domain-specific IgG, IgG1, IgG2a, IgG2b titers measured in biweekly serum samples. (Experiments were conducted independently in triplicates; all data are presented as mean ± SEM; n = 5–6 per group.) Two-way ANOVA with Sidak’s (B–E) and Tukey’s (F) multiple comparisons test were used for statistical analysis. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.