#THE NULL MICROSCOPY ANALYSIS!!!!!!!!!

[naivelocus]

Latent TGF-β binding protein 2 and 4 have essential overlapping functions in microfibril development

Ltbp1 and 3, but not 4, mRNA was abundantly expressed in the ciliary bodies of wild type and Ltbp2 null mice

Assuming that other LTBPs may compensate for the loss of LTBP-2 in stable microfibril formation in tissues other than the eyes in Ltbp2 null mice, we analyzed the expression of Ltbp1, 2, 3, and 4 mRNA in the lungs, aortae, and ciliary bodies of wild type and Ltbp2 null mice using RT-PCR (Fig. 1). Ltbp1 and 3 were expressed in ciliary bodies and in other tissues in both wild type and Ltbp2 null mice, indicating that LTBP-1 and -3 were not able to compensate for the loss of LTBP-2. On the other hand, only a trace amount of Ltbp4 was expressed in ciliary bodies, while it was abundantly expressed in the lungs and aortae, suggesting that LTBP-4 is a candidate molecule that could substitute for LTBP-2 to enable stable microfibril formation in tissues other than the eyes.

Figure 1: Expression of genes of the LTBP family in mouse tissues.

Gene expression of Ltbp1, 2, 3 and 4 and Gapdh in ciliary bodies, aorta and lung tissues from wild type (+/+) and Ltbp2−/− (−/−) mice was analyzed using RT-PCR. Expression of Ltbp4 was hardly detected in ciliary body tissues from either wild type or Ltbp2−/− mice.

Ltbp2/4S DKO mice had a higher mortality rate than that of Ltbp2 or 4S single KO mice

To assess whether LTBP-4 compensated for the loss of LTBP-2 in Ltbp2 null mice, we generated Ltbp2/4S DKO mice. Since Ltbp2 null mice do not have any embryonic and reproductive abnormalities, we set up an Ltbp2−/−; Ltbp4S+/− intercross to produce Ltbp2−/−; Ltbp4S−/− (Ltbp2/4S DKO) mice. Approximately 50% of Ltbp2/4S DKO mice did not survive until the weaning period (4 weeks after birth). We evaluated the survival rate of Ltbp2/4S DKO mice along with that of other genotypes in every week after birth (Supplementary Fig. S1a). Mice having a genotype of either Ltbp2−/−; Ltbp4S+/+ or Ltbp2−/−; Ltbp4S+/− did not show a major decrease in survival rate at 4 weeks after birth, and most of the Ltbp2/4S DKO mice survived by 3 weeks after birth. However, the survival rate of DKO mice suddenly decreased by 50% at 4 weeks after birth (Fig. 2). Dabovic et al., reported that Ltbp4S null mice show moderate lethality by 4 weeks after birth20. However, we analyzed the survival rate of Ltbp4S null mice at 4 weeks after birth and did not observe lethality (Fig. 2 and Supplementary Fig. S1b), indicating that juvenile lethality we found in our mutant mice was caused by the cooperative effect of LTBP-2 and LTBP-4 deficiencies. Interestingly, 70% of the Ltbp2/4S DKO mice that survived for 4 weeks after birth remained alive for 2 months or more after birth, suggesting a major critical period for survival of Ltbp2/4S DKO mice was between 3 weeks and 4 weeks after birth.

Figure 2: Survival rate of Ltbp2/4S mutant mice after birth.

Starting from postnatal day 7 (P7), pups of wild type, Ltbp4S−/− mice (produced by intercross of Ltbp4S+/− mice), Ltbp2−/− mice, and Ltbp2−/−; Ltbp4S−/− mice (produced by intercross of Ltbp2−/− and Ltbp4S+/− mice) were evaluated for survival every week. *P < 0.05, determined by log-rank test.

Vascular impairment in Ltbp2/4S DKO mice was similar to that found in Ltbp4S null mice

To understand the cause of increased lethality in Ltbp2/4S DKO mice, we analyzed the physical characteristics and tissue abnormalities in 8-week-old mice of all genotypes, including Ltbp2/4S DKO mice. Systolic, diastolic, and mean blood pressure were not different between the genotypes. However, the body weight of Ltbp2/4S DKO mice was significantly decreased compared with that of mice of other genotypes (Supplementary Fig. S2).

We reported that Ltbp4S null mice show aortic tortuosity at day 5 after birth19. We confirmed that neonatal mice of both Ltbp4S null and Ltbp2/4S DKO genotypes showed tortuous aorta (Supplementary Fig. S3a), although aortic tortuosity became less obvious in adult animals (8-week-old, Supplementary Fig. S3b). Mice with both wild type and Ltbp2 null genotypes showed no evident aortic abnormalities at the neonatal stage or the adult stage. Although microfibril defects in arteries were expected to result in ascending aortic aneurysm as occurs in Marfan syndrome caused by fibrillin-1 mutations, Ltbp2/4S DKO mice did not show apparent aneurysms.

Ltbp2/4S DKO mice show more severe emphysema than that found in Ltbp4S null mice

Ltbp4S null mice were reported to develop pulmonary emphysema18,21. Macroscopic examination confirmed obvious emphysema in Ltbp4S mice, and notably, Ltbp2/4S DKO mice showed a more severe form of emphysema than that found in Ltbp4S null mice (Fig. 3). Remarkably large terminal air sacs, indicative of disruption of alveolar walls, were observed in the lungs of Ltbp2/4S DKO mice (Fig. 3). We then conducted histological analysis of the lungs of wild type and mutant mice at different developmental stages. Lung development in mice proceeds after birth with terminal air sac septation and alveolarization occurring between P0.5 and P2120. Before birth, lung morphology was indistinguishable among wild type, Ltbp2 null, and Ltbp4S null mice, whereas the lungs of Ltbp2/4S DKO mouse had larger terminal air sacs, indicating a septation defect (Fig. 4a). At 8 weeks after birth, the lungs of wild type and Ltbp2 null mice showed normal development. However, airspace enlargement was observed in the lungs of adult Ltbp4S null mice, and these defects were even more evident in the lung of Ltbp2/4S DKO mice (Fig. 4b).

Figure 3: Gross morphology of Ltbp2/4S mutant lung in 8-week-old mice.

Mild and severe emphysematous phenotypes were observed in Ltbp4S−/− and Ltbp2/4S DKO mice, respectively.

Figure 4: Histological analysis of Ltbp2/4S mutant mouse lungs.

(a) Lung sections of mice at embryonic day 18.5 (E18.5) were stained with hematoxylin and eosin. Bars: 400 μm. (b) Lung sections of 8-week-old mice were stained with hematoxylin and eosin. Bars: 100 μm. The terminal air sac enlargement in the lungs of Ltbp2/4S DKO mice was more severe than that observed in the lungs of Ltbp4S−/− mice. (c) Electron micrographs of elastic fibers stained black by tannic acid in the lungs of 8-week-old mice. The elastic fibers of Ltbp4S−/− and Ltbp2/4S DKO lungs were disorganized. Note bare microfibril bundles (open arrow) and elastin deposits outside microfibrils (closed arrow) in Ltbp4S−/− lung tissues. In Ltbp2/4S DKO lung tissues, bare microfibril bundles were not observed, and elastic fibers were more severely fragmented. Bars: 1 μm. (d) Distribution of microfibrils in the lungs of 8-week-old mice. Cryosections of mouse lungs were stained with a rabbit anti-fibrillin-1 antibody and visualized using an Alexa-488-labeled anti-rabbit IgG antibody. Nuclei were stained with Hoechst 33258. Bars: 50 μm.

Degradation or impairment of elastic fibers is a major cause of emphysema22. Thus, we next analyzed elastic fibers in lung ECM, using electron microscopy. Elastic fibers were visualized by tannic acid staining. Continuous thick elastic fibers were observed along the alveolar walls in wild type and Ltbp2 null mice (Fig. 4c). These tannic acid-positive elastic fibers were fragmented as dot-like deposits in the lungs of Ltbp4S null mice (Fig. 4c). We also observed microfibril bundles devoid of elastin deposition in Ltbp4S null tissues, as previously reported21, indicating an essential function of LTBP-4 in mediating the proper assembly of elastin on microfibrils19. In Ltbp2/4S DKO lungs, we only observed smaller pieces of fragmented elastic fibers than in Ltbp4S null lung (Fig. 4c). Unlike Ltbp4S null tissue, we did not observe microfibril bundles without elastin deposition in Ltbp2/4S DKO tissue. Immunofluorescent staining of adult lungs using an anti-fibrillin-1 antibody revealed microfibrils in the wall of alveoli and blood vessels (Fig. 4d). Fibrillin microfibrils in the lungs of Ltbp2/4S DKO mice appeared discontinuous compared with those in lung tissues of the other genotypes. These results suggest that the severe emphysematous phenotype observed in Ltbp2/4S DKO mice could be attributable to degenerated elastic fibers with inefficient formation of bundled microfibrils, in addition to the terminal air sac septation defect.

Expression of elastogenic genes was enhanced at 3 weeks after birth but largely reduced at 8 weeks after birth in Ltbp2/4S DKO mouse lung tissue

Severe disruption of elastic fibers and alveolar walls in Ltbp2/4S DKO mouse lungs suggested impaired elastogenesis and/or an increase in elastolysis in these tissues. Therefore, we analyzed the mRNA expression level of the genes involved in elastic fiber formation and degradation in the lungs at different developmental stages. In neonatal lungs, the expression levels of ECM-related genes were not drastically changed, and the differences from wild type mice were less than 2-fold (Supplementary Fig. S4a). However, at 3 weeks after birth, when Ltbp2/4S DKO mice showed partial lethality (Fig. 2), the expression levels of Eln and Fbn2 were increased about 10-fold and 3-fold, respectively, in Ltbp2/4S DKO mouse lungs, compared with that of wild type lungs (Supplementary Fig. S4b). The increase in Fbn1 and Lox expression in Ltbp2/4S DKO tissues was not statistically significant, but these data indicate compensatory expression of Eln but no notable increase in elastolytic metalloproteases. These findings suggested that disruption of elastic fibers was primarily caused by defects in elastic fiber formation in Ltbp2/4S DKO lung tissues at this stage, rather than by degradation of elastic fibers. At 8 weeks after birth, which is the period when surviving Ltbp2/S DKO mice develop advanced-stage emphysema, Eln expression approached the normal level, and expression of other elastogenic genes, including Fn1, Fbn1, Fbn2, Fbln4, Fbln5, and Lox was dramatically decreased, whereas the expression level of Mmp12 (macrophage elastase) was increased nearly 3-fold (Supplementary Fig. S4c). These data suggest that elastic fiber-producing mesenchymal cells were impaired and that the MMP-12 produced by macrophages further degraded elastic fibers in Ltbp2/4S DKO mouse lung tissues at this stage.

Either LTBP-2 or -4 is necessary for microfibril formation in serum-free culture of mouse embryonic fibroblasts

To test the hypothesis that LTBP-4 compensates for the loss of LTBP-2 in microfibril formation, we analyzed microfibril formation in vitro using cultured mouse embryonic fibroblasts (MEFs) obtained from mutant mouse embryos. MEFs were cultured in serum-free media for 4 days and immunostained with anti-fibrillin-1 (for microfibril formation) as well as anti-LTBP-2 and -4 antibodies. Wild type MEFs produced a fibrillin-1–positive microfibril meshwork on the cells, and Ltbp2 null MEFs and Ltbp4S null MEFs showed slightly reduced signals by anti-fibrillin-1 antibody staining (Fig. 5a–c). Notably, microfibril meshwork formation was significantly abrogated in Ltbp2/4S DKO MEFs (Fig. 5d, and Supplementary Fig. S5a), but was rescued by the addition of either 5 nM of recombinant LTBP-2 or -4S (Fig. 5E,F, and Supplementary Fig. S5a). mRNA expression of Fbn1 and Fbn2 was even increased in Ltbp2/4S DKO MEFs by 2-fold compared to WT MEFs (Supplementary Fig. S5b). These data indicate that either LTBP-2 or LTBP-4S is required for fibrillin-positive microfibril meshwork formation under serum-free conditions. However, microfibril meshwork formation was indistinguishable between the genotypes when cells were cultured in 3% serum-containing medium (Supplementary Fig. S6), suggesting that an unknown element(s) contained in fetal bovine serum promoted microfibril formation without a requirement for endogenous LTBPs. In contrast, elastin deposition onto microfibrils, which was visualized by immunostaining with anti-elastin and anti-fibulin-5 antibodies after cells were cultured in 10% serum-containing media for 14 days, strictly depended on the presence of LTBP-4S protein, with or without LTBP-2 (Supplementary Figs S7 and S8).

Figure 5: Microfibril formation on cultured MEFs from mutant mice.

MEFs of all genotypes were cultured in serum-free media for four days, and the ECM was stained with anti-fibrillin-1, anti-LTBP-2, and anti-LTBP-4 antibodies, followed by fluorophore-labeled secondary antibodies corresponding to each first antibody (green for fibrillin-1, white for LTBP-2, and red for LTBP-4). Nuclei were stained with Hoechst 33258. MEFs lacking either LTBP-2 (b) or LTBP-4S (c) produced a weak but clear fibrillin-1 meshwork as observed in wild type MEFs (a), whereas MEFs lacking both LTBP-2 and LTBP-4S only produced aggregated deposition but no fibrous fibrillin-1–positive meshwork (d). Supplementation of recombinant protein of LTBP-2 (e) or LTBP-4S (f) at a concentration of 5 nM into the cultured medium restored microfibril meshwork formation on Ltbp2/4S DKO MEFs. Bars: 150 μm.

Disruption of ciliary zonules in Ltbp2 null mice was compensated by ectopic LTBP-4 overexpression

Next, we overexpressed LTBP-4 in the eyes of mice with an Ltbp2 null background, to assess whether LTBP-4 could compensate for the loss of LTBP-2 in vivo. Knock-in mice that had a CAG-promoter-driven Ltbp4S cDNA at the Rosa26 locus displayed enhanced expression of Ltbp4S in the whole body, and survived to adulthood without any gross malformation (Supplementary Fig. S9). These mice were crossed with Ltbp2 null mice to obtain Ltbp2 null; Ltbp4S KI mice. Without the Ltbp4S knock-in allele, eyes from 4-week-old Ltbp2 null mice exhibited fragmented ciliary zonules (Fig. 6a,b). However, the zonules were partially restored by the presence of ectopically expressed LTBP-4S in the eyes of Ltbp2 null; Ltbp4S KI mice (Fig. 6c), while LTBP-4S protein co-localized with the fibrillin-1 present in ciliary zonule microfibrils. These data indicated that LTBP-2 and -4 have an overlapping functions in forming the robust architecture of ciliary zonules in vivo.

Figure 6: Restoration of ciliary zonule formation in Ltbp2−/−; Ltbp4S KI mice.

Ltbp4S KI mice ubiquitously expressed LTBP-4S in the whole body, including ciliary bodies, whereas LTBP-4 was not expressed in ciliary bodies of wild type mice. By crossing these mice with Ltbp2−/− mice, Ltbp2−/−; Ltbp4S KI mice were produced, and ciliary zonule formation of the mouse eyes was visualized using immunofluorescent analysis, as in Fig. 5. LTBP-2 and fibrillin-1 co-localized on ciliary zonules of wild type mouse eyes, but fibrous fibrillin-1 staining was largely absent in the Ltbp2−/− eye. Ectopic expression of LTBP-4 in ciliary bodies restored ciliary zonule formation in Ltbp2−/−; Ltbp4S KI mouse eyes. Bars: 100 μm.

— Nature Scientific Reports