Abstract
Gap junction beta-1 (GJB1) gene mutations affecting the gap junction protein connexin32 (Cx32) cause the X-linked Charcot-Marie-Tooth disease (CMT1X), a common inherited neuropathy. Targeted expression of virally delivered Cx32 in Schwann cells following intrathecal injection of lentiviral vectors in the Cx32 knockout (KO) mouse model of the disease has led to morphological and functional improvement. To examine whether this approach could be effective in CMT1X patients expressing different Cx32 mutants, we treated transgenic Cx32 KO mice expressing the T55I, R75W or N175D CMT1X mutations. All three mutants were localized in the perinuclear compartment of myelinating Schwann cells consistent with retention in the ER (T55I) or Golgi (R75W, N175D) and loss of physiological expression in the non-compact myelin. Following intrathecal delivery of the GJB1 gene we detected the virally delivered wild-type (WT) Cx32 in non-compact myelin of T55I KO mice, but only rarely in N175D KO or R75W KO mice, suggesting dominant-negative effects of the R75W and N175D mutants but not of the T55I mutant on co-expressed WT Cx32. GJB1 treated T55I KO mice showed improved motor performance, lower ratios of abnormally myelinated fibers and reduction of inflammatory cells in spinal roots and peripheral nerves compared with mock-treated littermates. Either partial (N175D KO) or no (R75W KO) improvement was observed in the other two mutant lines. Thus, certain CMT1X mutants may interfere with gene addition therapy for CMT1X. Whereas gene addition can be used for non-interfering CMT1X mutations, further studies will be needed to develop treatments for patients harboring interfering mutations.
Introduction
The X-linked Charcot-Marie Tooth (CMT1X) disease is one of the commonest inherited neuropathies, characterized by progressive weakness and atrophy of distal limb muscles, loss of reflexes, sensory loss and reduced nerve conduction velocities. CMT1X results from mutations in the gap junction beta-1 (GJB1) gene, which encodes the gap junction (GJ) protein connexin32 (Cx32) (1–3). Cx32 forms intracellular GJ channels through the non-compact myelin layers, at paranodal loops and Schmidt–Lanterman incisures of Schwann cells (4,5). These channels serve important homeostatic and axon-glial signaling functions in peripheral myelinated fibers. There are over 400 different GJB1 mutations reported so far (http://hihg.med.miami.edu/code/http/cmt/public_html/index.html#/; date last accessed February 22, 2018) affecting all domains of Cx32, even the non-coding gene regions (6,7). In addition to peripheral neuropathy, a number of CMT1X mutations also result in central nervous system (CNS) phenotypes characterized by spasticity, hyperactive reflexes, extensor plantar responses, ataxia, or acute reversible encephalopathy (5,8–11).
In vitro studies using HeLa cells have revealed different cellular expression patterns of CMT1X mutants and variable ability to form GJs (12–16). Some of the mutants may form GJ-like plaques, which are areas of connexin immunoreactivity on the cell membrane representing an accumulation of many GJ channels, and carboxy-terminus mutants may even retain some functional properties (17,18). However, most of the Cx32 mutations examined so far have shown abnormal intracellular trafficking with failure to form GJ channels on the cell membrane and retention of the mutant Cx32 protein either in the endoplasmic reticulum (ER) or in the Golgi apparatus (13,14,16). Intracellularly retained mutants are degraded by the proteasomes and/or the lysosomes (19).
CMT1X mutants that have been expressed in vivo have shown similar cellular expression patterns as they did in vitro (20,21). These in vivo models have also clarified that Cx32 mutants have no toxic effects and lead to loss of function regardless of cellular localization with only rare exceptions (22). This is in keeping with clinical studies demonstrating that CMT1X patients show a similar phenotype and disease progression regardless of the type of mutation (23), including complete deletions and non-coding region mutations (7).
Although not relevant for most CMT1X patients as only one GJB1 allele is expressed in each cell, dominant-negative effects of certain Cx32 mutants on the non-physiologically co-expressed WT Cx32 have been shown experimentally and need to be considered in the setting of a possible gene addition therapy. In vitro co-expression of certain Cx32 mutants and WT Cx32 resulted in impairment of WT Cx32 function (24,25). Moreover, direct interactions between certain mutants and WT Cx32 causing intracellular retention of the WT protein have been demonstrated (14). Dominant-negative effects were observed with Golgi-retained but not with ER-retained mutants. This is in accordance with in vivo data showing that transgenic mice expressing the ER-retained T55I mutant on WT background develop no nerve pathology, while mice expressing the Golgi-retained R75W or R142W mutants develop some pathology even on a WT background, indicating dominant-negative effects on the co-expressed WT protein (20,21).
Based on our recent findings that intrathecal delivery of GJB1 driven by the Schwann cell-specific myelin protein zero (Mpz) promoter in a lentiviral vector leads to Cx32 expression and improved nerve pathology in the Cx32 knockout (KO) (null) mouse (26,27), we further tested this approach in mice expressing three different CMT1X mutations on a KO background, the ER-retained T55I, or the Golgi-retained R75W and N175D. We show here that intrathecal gene delivery resulted in normal expression of virally delivered Cx32 in T55I KO mice and improvement of the phenotype. In contrast, R75W mutants and to a lesser degree N175D mutants showed mostly perinuclear Cx32 retention and lack of functional or morphological improvement. Our results confirm the efficacy of gene addition therapy to treat CMT1X but indicate that in patients expressing certain interfering mutations alternative approaches may be needed.
Results
Expression of the virally delivered human GJB1 gene in CMT1X mutant mice
In previous studies, we have shown that Schwann cell specific expression of WT Cx32 delivered by the LV.Mpz.GJB1 lentiviral vector resulted in improvement of morphological and functional properties of the peripheral nerves in Cx32 KO mice following intraneural (27) or lumbar intrathecal injection (26). To further study the potential of this gene therapy approach to treat CMT1X patients harboring different GJB1 mutations, we injected the same vector in three different transgenic mouse models expressing on a Cx32 KO background representative Cx32 mutants that are either retained in the ER (T55I) or in the Golgi (R75W, N175D). The generation and characterization of these transgenic lines is described in the Materials and Methods and in Supplementary Material (Supplementary Material, Fig. S1). Furthermore, immunoblot analysis was used to confirm the expression levels of reporter gene enhanced green fluorescent protein (EGFP) and mutant Cx32 in the N175D KO mutant line compared with the other two previously generated lines, T55I KO and R75W KO (21). EGFP expression was similar in T55I KO, R75W KO and N175D KO mice and comparable to the levels of the previously generated R75W mutant line (Fig. 1A). Mutant Cx32 was also detected in all three lines (Fig. 1B and C), but levels were slightly higher in tissues of the Golgi-retained R75W and N175D mutants compared with the ER-retained T55I mutant (Fig. 1D).
Immunoblot analysis of EGFP and Cx32 expression levels in sciatic nerve samples from transgenic lines expressing CMT1X mutants on a Cx32 KO background used for gene therapy. (A) Immunoblot for EGFP shows expression in T55I/Cx32 KO (KO T), R75W/Cx32KO (KO R) and N175D/Cx32 KO (KO N) mice at comparable levels, similar to a previously generated R75W mutant line (TG+) shown as positive control, whereas EGFP is absent from a simple Cx32 KO mouse sample (KO). Immunoblot analysis of Cx32 expression levels using a rabbit-anti Cx32 antiserum (B) or a mouse monoclonal antibody (C) reveals expression of mutant Cx32 at higher levels in the case of Golgi-retained R75W and even more in N175D mutants compared with the ER-retained T55I mutant. The Cx32 KO sample used as negative control shows a non-specific band only with the rabbit antibody. Wild type (WT) nerve sample is used as positive control. Both tubulin immunoblots as well as Coomassie stained gels revealing the characteristic myelin protein zero (P0) band are shown as loading controls under the EGFP and Cx32 blots. (D) Quantification of normalized Cx32 levels (n = 3 different mice from each genotype) using the mouse anti-Cx32 antibody shows similar levels of the N175D mutant compared with WT Cx32, while the R75W and T55I mutants show lower expression levels.
We first delivered the LV.Mpz.Egfp (mock) or the LV.Mpz.GJB1 (full) vectors intrathecally in 2-month-old T55I, R75W and N175D Cx32 KO mice and analyzed the expression of Cx32. Sciatic nerve teased fibers and lumbar spinal roots were immunostained 4 and 8 weeks post-injection with antibodies against Cx32 and Contactin associated protein (Caspr) (a marker for paranodal axonal domains). In tissues from mock vector injected T55I KO mice, Cx32 was detected only in the perinuclear area of Schwann cells (Fig. 2A and G). In contrast, in T55I KO mutants injected with the LV.Mpz.GJB1 vector, Cx32 was not only detected in the perinuclear Schwann cell cytoplasm, but also in non-compact myelin (incisures and paranodes) in both lumbar roots (Fig. 2D) and sciatic nerves (Fig. 2J). We interpret these findings as showing that WT Cx32, from the vector, was normally localized and did not interact with the T55I mutant that was retained in the ER. Expression of virally delivered WT Cx32 in T55I KO mice was also confirmed by double staining with EGFP and Kv1.1. Staining with Cx32 and EGFP showed that WT Cx32 was detected in the paranodal areas in fully treated (Supplementary Material, Fig. S2B) but not in mock-treated T55I KO mice (Supplementary Material, Fig. S2A). Furthermore, double staining with the juxtaparanodal marker Kv1.1 confirmed the correct localization of WT Cx32 (Supplementary Material, Fig. S2C). Paranodal expression of WT Cx32 was also detected in femoral nerve teased fibers double immunostained with axonal domain markers (Supplementary Material, Fig. S2D and E).
Expression of Cx32 in Schwann cells in PNS tissues of three different Cx32 mutant mouse models following intrathecal vector delivery. These are images of anterior lumbar root sections (A–F) or sciatic nerve teased fibers (G–L) examined 8 weeks following LV.Mpz-Egfp (mock) or LV.Mpz-GJB1 (full) vector intrathecal injection, immunostained for Cx32 (red) and paranodal marker Caspr (green). Cell nuclei are stained with DAPI (blue). The spinal roots of T55I KO (A), R75W KO (B) and N175D KO (C) (mock vector) injected mice show baseline Cx32 (red) expression only in perinuclear Schwann cells cytoplasm (shown in insets) but not in paranodal areas labeled with Caspr. Following injection of the full vector, there is in addition to perinuclear localization (open arrowheads in insets) also expression of virally delivered Cx32 in the paranodal areas (arrows) surrounding the paranodal marker Caspr only in spinal roots of the T55I KO (D) and in some cases of N175D KO (F) mice, but not in the R75W KO (E). Paranodal areas are shown in separate and merged channels under the overview images. Likewise, sciatic nerve teased fibers from T55I KO (G), R75W KO (H) and N175D KO (I) mock injected mice show only perinuclear expression of Cx32 in all three transgenic lines (open arrowheads), whereas LV.Mpz-GJB1 injected mutants show additional expression of Cx32 in the paranodal areas (arrows) only in the T55I KO (J) and in some cases in the N175D KO (L) but not in the R75W KO (K) mouse fibers. Scale bars in F, L: 10 μm. Quantification of Cx32 positive Caspr-labelled nodal areas showed that both in the lumbar roots (M) and in sciatic nerve fibers (N) the rates of correct paranodal expression of virally delivered Cx32 reached ∼50% in the T55I KO mutant tissues while these ratios were significantly lower in the N175D KO and even more in R75W KO mutants (Numbers see text; P-values obtained with Student’s t-test).
In contrast to T55I KO mice, in R75W KO (Fig. 2E and K) and N175D KO (Fig. 2F and L) mice injected with the LV.Mpz.GJB1 vector, Cx32 was only rarely found in paranodal areas and was mostly detected in the perinuclear areas of Schwann cells as in mock-treated mice (Fig. 2B, C, H and I). This result indicated that the R75W and N175D mutants resulted in the retention of the (virally delivered) WT Cx32 in the perinuclear region, such that it did not reach the paranodes.
We quantified the Cx32-positive paranodes in teased fibers from the anterior lumbar roots (Fig. 2M) and sciatic nerves (Fig. 2N) in the different mutants injected with the LV.Mpz-GJB1 vector. In treated T55I KO mice, the % paranodes that were immunoreactive for Cx32 was 51.8 ± 11.7% in spinal roots and 52.9 ± 4.5% in sciatic nerve fibers (n = 5 mice examined). In treated R75W KO mice, only 6.4 ± 4.1% and 6.5 ± 10.1% of paranodes were Cx32-positive in anterior roots and sciatic nerves, respectively (n = 6). In treated N175D KO mice, 16.8 ± 7.6% and 17.2 ± 5.7% were Cx32-positive in spinal roots and sciatic nerves, respectively (n = 5) (Fig. 2M and N). Thus, while in T55I KO mice the expression rates of virally delivered Cx32 were comparable to Cx32 KO mice (26), R75W KO mice and to a lesser degree the N175D mice showed significantly reduced rates of Cx32 expression in non-compact myelin areas, suggesting an interference of the Golgi-retained mutants with virally delivered WT Cx32 preventing its normal trafficking.
Improved motor function in certain CMT1X mutants after gene therapy
In order to assess whether gene therapy improves motor performance in CMT1X mutant mice, we performed behavioral analysis at 4 and 8 months of age (2 and 6 months post-injection). We used rotarod and foot grip analysis in LV.Mpz.GJB1 (full vector) compared with LV.Mpz.Egfp (mock) treated groups of mice from all three genotypes. In treated T55I KO mice, behavioral tests showed improvement of the motor performance compared with the mock-treated littermates. At 4 months of age (Fig. 3A), mice treated with the full vector (n = 14) remained longer on the rotarod at the speed of 32 RPM 168.8 ± 174.9 s compared with 81.5 ± 118.3 s in mock treated (n = 17) (Fig. 3A;Supplementary Material, Table S1; P = 0.0209), whereas at a lower speed of 20 RPM the difference between the groups was not significant (360.0 ± 184.9 s in fully treated compared with 241.9 ± 196.5 s in mock treated, P > 0.05). Similar results were obtained at 8 months (Fig. 3D;Supplementary Material, Table S1): at 20 RPM fully treated mice remained for 458.2 ± 135.6 s (n = 13) whereas mock-treated for 347.2 ± 181.4 s (n = 13) (P > 0.05); at 32 RPM fully treated remained for 226.7 ± 176.2 s whereas mock treated 83.7 ± 72.7 s (P = 0.0102).
Behavioral analysis of LV.Mpz-GJB1-injected compared with mock injected CMT1X mutant mice. These are the results of rotarod (A–F) and foot grip (G–I) testing of motor performance in LV.Mpz-GJB1 treated (GJB1) compared with mock treated mutant mice, as indicated. Rotarod analysis showed improved motor performance at the higher speed of 32 rotations per minute (RPM) but not at 20 RPM in fully treated T55I KO mice compared with the mock treated at 4 months (n = 14 full and n = 17 mock) (A) and at 8 months of age (n = 13 full and n = 13 mock) (D). Foot grip analysis showed also improvement of the fully treated compared with the mock-treated T55I KO mice at both 4 and 8 months of age (G). In contrast, rotarod analysis showed no improvement of motor performance in fully treated R75W KO mice compared with the mock treated at both 4 (n = 23 full and n = 11 mock) (B) and 8 months (n = 14 full and n = 8 mock) (E) of age at the two speeds tested. Likewise, foot grip analysis showed no improvement of fully treated compared with mock-treated R75W KO mice at both age groups (H). Finally, similar to R75W KO mice, rotarod analysis (C, F) showed lack of improvement in fully treated (n = 13) compared with the mock treated (n = 13) N175D KO mice, while an improvement was detected by the foot grip analysis (I) at both 4 (n = 13 full and n = 13 mock) and 8 months (n = 14 full and n = 13 mock).
Improved motor performance in treated T55I KO mice was also demonstrated by the foot grip test (Fig. 3G;Supplementary Material, Table S1), in which force generated by the hind limbs was increased compared with mock treated littermates. At 4 months of age, force reached 104.6 ± 38.8 g in fully treated and 70.6 ± 24.5 g in mock treated mice (P = 0.0131), while at 8 months force reached 121.7 ± 58.8 g in fully treated and 39.3 ± 16.1 g in mock treated mice (P < 0.0001).
In contrast to the T55I KO mutant line, either slight or no improvement was observed in the motor performance of the other two mutant lines treated with LV.Mpz.GJB1 injection. In R75W KO mice, no improvement was observed in either rotarod or foot grip analysis in fully treated compared with mock treated littermates (n = 8–23 per group) both at 4 or at 8 months (Fig. 3B, E and H;Supplementary Material, Table S1). By comparison, treated N175D KO mutant mice (n = 13–14 per group) showed a mild improvement that was only evident in the foot grip and not in the rotarod analysis. Both at 4 and 8 months of age fully treated N175D KO mice showed no significant difference compared with mock treated mice in rotarod performance at both speeds tested (Fig. 3C and F;Supplementary Material, Table S1). However, foot grip analysis (Fig. 3I;Supplementary Material, Table S1) showed a significant improvement at both 4 and 8 months with force values reaching 122.6 ± 41.7 g in fully treated compared with 61.8 ± 23.8 g in mock treated N175D KO mice at 4 months (P < 0.0001) and 151.5 ± 64.5 (full) versus 62.9 ± 38.1 (mock) at 8 months (P < 0.0001).
Different outcomes of morphological analysis in treated CMT1X mutant mice
Transverse, semithin sections of lumbar spinal cord with attached anterior and posterior roots, as well as femoral motor and mid-sciatic nerves, were prepared from 8-month-old fully treated and mock-treated littermates (n = 8–14 mice per treatment group) from T55I, R75W and N175D KO mutant lines. Morphometric analysis of semithin sections was performed by an observer who was blinded to the treatment condition. Multiple roots, as well as bilateral nerves, were examined, and results were averaged per mouse. The number of abnormally myelinated fibers, including demyelinated and remyelinated fibers, was counted and their proportion to the total number of fibers calculated (21,28). Foamy macrophages were also counted (27), and their numbers per 1000 myelinated fibers were compared (to account for variations in root and nerve size).
In T55I KO mutants, the ratios of abnormally myelinated fibers as well as the numbers of foamy macrophages were reduced in fully treated (n = 12) compared with the mock treated littermates (n = 12) in all tissues examined (Supplementary Material, Table S2). In anterior lumbar spinal roots (Fig. 4A–F) in fully treated mice the ratio of abnormal fibers was 0.073 ± 0.02 whereas in mock treated mice this ratio was 0.153 ± 0.03 (Fig. 4E;P < 0.0001, Mann–Whitney U test). Likewise, macrophage numbers were reduced in anterior roots (4.371 ± 2.33/1000 fibers) from fully treated mice compared with those from mock treated mice (9.502 ± 2.72/1000 fibers) (Fig. 4F;P < 0.0001).
Morphological analysis of CMT1X mutant mouse spinal roots following intrathecal delivery of the LV.Mpz-GJB1 vector. Representative images of semithin sections of anterior lumbar spinal roots attached to the spinal cord at low (left column) and higher (middle column) magnification, as well as morphometric analysis results (right column) from mock or full (GJB1) vector treated mutant mice as indicated, at 8 months of age (6 months after treatment). (A–F) Improved pathology in lumbar roots of treated T55I KO mice. LV.Mpz-GJB1 injected T55I KO mice (B, D) show improved myelination compared with mock-treated littermates (A, C) with fewer demyelinated (*) and remyelinated (r) fibers. Quantification of the ratios of abnormally myelinated fibers in multiple roots (n = 13 mice per group) confirms significant improvement (E) as well as significant reduction in the numbers of foamy macrophages (F) in the fully treated compared with mock treated littermates. (G–L) In contrast, no pathology rescue is found in treated R75W KO mouse spinal roots (H, J), which show similar abnormalities as mock treated littermates (G, I) with many demyelinated and remyelinated fibers present. Quantification of pathology (n = 8 mice per group) confirms the lack of improvement with similar ratios of abnormally myelinated fibers (K) and numbers of foamy macrophages (L) in treated compared with mock treated R75W KO littermates. (M–R) N175D KO mice show only partial improvement of root pathology (N, P) after treatment compared with mock-treated littermates (M, O). Quantification (n = 13 mice per group) confirms a modest improvement in the rate of abnormally myelinated fibers in treated compared with mock-treated N175D KO mice (Q), but not in the range seen in T55I KO mice, while the number of macrophages shows no significant improvement (R). Scale bars shown in panels N and P.
Morphometric analysis of mid-sciatic nerves (Fig. 5A–F) showed significant improvement in all pathological parameters, including the ratio of abnormal fibers in fully treated (0.0461 ± 0.012) compared with mock-treated T55I KO mice (0.1018 ± 0.017; P < 0.0001) (Fig. 5E) and macrophage numbers (mock-treated: 7.709 ± 2.30/1000 fibers; fully treated 3.701 ± 1.84/1000 fibers; P < 0.0001) (Fig. 5F). Improvement was also observed in the femoral motor nerves (Fig. 6A–F) both in the ratio of abnormal fibers (mock-treated 0.186 ± 0.039; fully treated 0.087 ± 0.010; P < 0.0001) (Fig. 6E) as well as in the numbers of macrophages/1000 fibers (mock-treated: 10.616 ± 2.59/1000 fibers; fully treated 4.216 ± 1.33/1000 fibers; P < 0.0001) (Fig. 6F).
Alterations in myelination of sciatic nerve fibers CMT1X mutant mice following intrathecal gene therapy with the LV.Mpz-GJB1 vector. Representative images of semithin sections of mid-sciatic nerves at low (left column) and higher (middle column) magnification, as well as morphometric analysis results (right column) from mock or full (GJB1) vector treated mutant mice as indicated, at 8 months of age (6 months after treatment). (A–F) Improved pathology in sciatic nerve fibers of T55I KO mice with fewer demyelinated (*) or remyelinated (r) fibers in LV.Mpz-GJB1 injected (B, D) compared with LV.Mpz-mock injected littermates (A, C). Quantification of the ratios of abnormally myelinated fibers in multiple sciatic nerves (n = 13 mice per group) confirms significant improvement (E) as well as significant reduction in foamy macrophages numbers (F) in treated compared with mock treated littermates. (G–L) Lack of pathology rescue in treated R75W KO mice with similar numbers of demyelinated and remyelinated fibers in treated (H, J) compared with mock treated (G, I) animals. Quantification of the ratios of abnormally myelinated fibers (K) in multiple sciatic nerves (at least n = 8 mice per group) confirms the lack of improvement, as well as unchanged numbers of foamy macrophages (L) in fully compared with the mock treated littermates. (M–R) N175D KO mice show partial and modest improvement of sciatic nerve pathology in treated (N, P) compared with mock-treated littermates (M, O) with reduced numbers of abnormal fibers. Quantification of the ratios of abnormally myelinated fibers (Q) in multiple sciatic nerves (at least n = 13 mice per group) confirms an improvement in the numbers of abnormally myelinated fibers whereas the numbers of foamy macrophages (R) did not change significantly. Scale bars shown in panels N and P.
Alterations in myelination of femoral motor nerve fibers in following intrathecal gene therapy with the LV.Mpz-GJB1 vector in CMT1X mutant mice. Representative images of semithin sections of femoral motor branches at low (left column) and higher (middle column) magnification, as well as morphometric analysis results (right column) from mock or full (GJB1) vector treated mutant mice as indicated, at 8 months of age (6 months after treatment). (A–F) Improved pathology in femoral motor nerves of treated (B, D) compared with mock-treated (A, C) T55I KO mice with fewer demyelinated and remyelinated fibers. Quantification of the ratios of abnormally myelinated fibers in multiple femoral nerves (n = 13 mice per group) confirms significant improvement (E) as well as reduction in foamy macrophages numbers (F) in the fully treated compared with mock-treated littermates. (G–L) In contrast, there is no pathology rescue in treated R75W KO mice (H, J) compared with mock treated (G, I) littermates. Quantification of the ratios of abnormally myelinated fibers (K) in multiple femoral nerves (at least n = 8 mice per group) confirms the lack of improvement as well as unchanged numbers of foamy macrophages (L). (M–R) Likewise, there is no pathology rescue in femoral nerves of N175D KO mice. Representative images show similar abnormalities in treated (N, P) as in mock-treated (M, O) littermates. Quantification of the ratios of abnormally myelinated fibers in N175D KO femoral nerves (at least n = 13 mice per group) shows lack of improvement in the ratios of abnormally myelinated fibers (Q) and in the numbers of foamy macrophages (R) in fully compared with mock-treated littermates. Scale bars shown in panels N and P.
In contrast to T55I KO mice, in R75W KO mutant mice there was no significant improvement of pathology in any of the tissues examined in fully treated (n = 11) compared with mock treated littermates (n = 8) (Supplementary Material, Table S3). In anterior lumbar spinal roots (Fig. 4G–L) the ratio of abnormal fibers was 0.162 ± 0.06 in fully treated compared with 0.152 ± 0.06 in mock treated mice (Fig. 4K;P > 0.05). Macrophage numbers in anterior roots were 6.6 ± 3.33/1000 fibers in fully treated compared with 8.5 ± 6.15/1000 fibers in mock treated (Fig. 4L;P > 0.05). Likewise, in mid-sciatic nerves (Fig. 5G–L) the ratio of abnormal fibers in fully treated R75W KO mice was 0.087 ± 0.03 compared with 0.076 ± 0.01 in mock-treated mice (Fig. 5K;P > 0.05). Macrophages in mock-treated mice were 3.5 ± 1.06/1000 fibers compared with 4.4 ± 2.10/1000 fibers in fully treated mice (Fig. 5L;P > 0.05). Finally, in femoral motor nerves (Fig. 6G–L) the ratio of abnormal fibers was 0.254 ± 0.03 in mock and 0.275 ± 0.07 in fully treated mice (Fig. 6K;P > 0.05), while macrophages numbers were 5.6 ± 2.22/1000 fibers in mock and 7.7 ± 1.79/1000 fibers in fully treated mice (Fig. 6L;P > 0.05).
Finally, in N175D KO mutant mice morphological analysis indicated a slight improvement only in lumbar roots and sciatic nerves but not in femoral nerves (Supplementary Material, Table S4). In anterior lumbar spinal roots (Fig. 4M–R) of fully treated mice (n = 14) the ratio of abnormal fibers was 0.11 ± 0.05 whereas in mock treated littermates (n = 13) this ratio was 0.153 ± 0.05 (Fig. 4Q;P = 0.0291). Macrophage numbers in anterior roots were 7.3 ± 4.90/1000 fibers in fully treated compared with 10.3 ± 4.76/1000 fibers in mock treated mice (Fig. 4R;P > 0.05). In mid-sciatic nerves of N175D KO mice (Fig. 5M–R) there was an improvement in the ratio of abnormal fibers in fully treated (0.064 ± 0.02) compared with mock-treated mice (0.082 ± 0.02; P = 0.0145) (Fig. 5Q) but no significant improvement in the macrophage numbers (4.7 ± 1.84/1000 fibers in mock-treated; 4.0 ± 1.62/1000 fibers in fully treated; P > 0.05) (Fig. 5R). Femoral motor nerves (Fig. 6M–R) showed no significant improvement in either the ratio of abnormal fibers (mock-treated 0.20 ± 0.04; fully treated 0.16 ± 0.08; P > 0.05) (Fig. 6Q) or in the number of macrophages/1000 fibers (mock-treated 7.7 ± 1.28/1000; fully treated 6.4 ± 2.71/1000; P > 0.05) (Fig. 6R).
Discussion
We have performed gene addition therapy using a lentiviral vector in three relevant models of CMT1X. A single intrathecal injection of a lentiviral vector carrying the human GJB1 under the control of the Mpz promoter resulted in Schwann cell-specific expression, an approach previously demonstrated in Cx32 KO mice (26,27). Virally delivered WT Cx32 was correctly expressed in PNS tissues of the T55I KO mutants but not in the tissues of the R75W KO and minimally in N175D KO mutants. Consequently, functional and pathological rescue similar to Cx32 KO mice was demonstrated only for the T55I KO mice, whereas the other two lines showed either minimal (N175D) or no (R75W KO) therapeutic benefit. The dominant-negative interaction is not relevant to the pathogenesis of CMT1X, because cells only express one GJB1 allele, but could be a major hurdle for potential gene addition therapy.
We and others have shown that many Cx32 mutants associated with CMT1X are retained in the ER or the Golgi, and do not reach the cell membrane in order to form GJ plaques (13,14,16,25). When co-expressed with WT Cx32 in vitro or in vivo, ER retained mutants did not interact with the WT Cx32, while some Golgi-retained mutants impaired partially or completely the correct localization and function of WT Cx32 (14,21,25). Based on these data, our aim was to rescue the demyelination caused by the loss of normal Cx32 function by intrathecal delivery of the WT Cx32 in the presence of three representative Cx32 mutants associated with CMT1X.
In order to proceed with the gene therapy trial in all CMT1X mutant mice we had to first characterize the newly established N175D KO transgenic line. This is the first time that a mouse model for the N175D mutation has been generated. Previous in vitro experiments showed that this mutant N175D Cx32 is mainly localized in the Golgi and fails to form functional GJ plaques (14,29) similar to other Golgi-retained mutants including the R75W (16). Even small changes that cause loss of E2-E2 hydrogen bonds can modify the properties of the N175D mutant protein (29). Generation of the mouse model showed that the N175D mutant is expressed by myelinating cells both in CNS and PNS. Confirming the in vitro results, it is localized in the perinuclear areas without expression in non-compact myelin areas in peripheral nerves, and without formation of GJ-like plaques, similar to the in vivo localization of the R75W mutant (21). Pathological analysis in untreated N175D KO mice (mock group) revealed a similar degree and progression of pathology as described previously in Cx32-null mice (30,31) and in other mutant lines (21). The baseline phenotype of N175D KO mice further supports the view that most CMT1X mutants have no toxic effects when expressed alone and that the neuropathy results from loss of normal Cx32 function.
In previous studies, we have demonstrated that intrathecal delivery of human Cx32 can rescue the demyelinating neuropathy in the Cx32 KO (Gjb1-null) mouse model of CMT1X (26). This is the first attempt to use a gene addition approach for the treatment of CMT1X models additionally expressing Cx32 mutations. Using this approach we achieved expression of the WT human Cx32 in the paranodal areas mostly in the T55I KO mutant, but only partially in the N175D KO and minimally in the R75W KO mutants. Virally delivered Cx32 was normally localized at the paranodal myelin areas of teased fibers and lumbar roots of the T55I mutants while the T55I mutant was present at the perinuclear areas. This is an indication that the endogenous mutant Cx32 did not interact with the virally delivered WT Cx32. Moreover, behavioral motor testing including rotarod analysis at higher speeds and foot grip analysis confirmed significant improvement in fully treated compared with mock treated mice. Finally, morphological analysis results confirmed improved myelination and reduced inflammation in peripheral nerves and roots. These results are in keeping with the findings in T55I transgenic mice (21) showing that the T55I mutant does not cause any neuropathy on a WT background, suggesting that it does not exhibit dominant-negative effects on endogenously expressed mouse WT Cx32. This is also in accordance with our in vitro results showing no direct interaction between the two proteins when co-expressed in HeLa cells and no functional impairment of WT Cx32 in the presence of the T55I mutant (14).
In contrast to the T55I mutant, in the case of the N175D KO mutant gene addition resulted only in partial and modest improvement evident in both functional and morphological outcomes. This is an indication that the N175D mutation exerts partial dominant-negative effects on the virally delivered WT Cx32, confirming our recent in vitro study showing that the N175D mutant interacts directly with WT Cx32 and impairs partially the expression of WT Cx32 causing reduced formation of functional GJ plaques (14). The presence and dominant-negative effects of the N175D mutant disrupted completely the expression and therapeutic effect of the WT Cx32 in the motor fibers of the femoral nerve and partially in the lumbar roots and sciatic nerves as indicated by the measurements of the abnormally myelinating fibers and macrophages. This could be owing to the fact that our viral vector is integrated at lower levels in femoral nerves that in the other two tissues examined (26). Furthermore, femoral motor nerves show the highest degree of pathology in CMT1X models (21,26,31) making it more difficult to achieve a therapeutic benefit. Finally, even in the sciatic nerves and lumbar roots modest improvement was observed only in the numbers of the abnormally myelinating fibers and not the numbers of macrophages. Thus, although WT Cx32 is expressed, expression levels are not high enough to overcome the dominant-negative effects of the N175D mutant.
The R75W mutation proved to have the strongest dominant-negative effect on the expression of the virally delivered WT Cx32. This resulted in a complete loss of therapeutic benefit following gene therapy with no improvement observed either in motor behavioral analysis or in the morphological analysis of PNS pathology. This implies that the R75W has stronger dominant-negative effects compared with the N175D mutation although both are Golgi-retained mutants. This is in accordance with co-immunoprecipitation experiments in vitro showing that the R75W mutant has the strongest direct interaction with co-expressed WT Cx32 (14). Moreover, R75W reduced the formation of functional GJ plaques by WT Cx32 in HeLa cells and disrupted GJ connectivity more severely than the N175D mutant. In contrast to T55I, the R75W mutant was also shown to cause peripheral nerve pathology on a WT background (21) indicating strong dominant-negative effects even on the endogenously expressed mouse Cx32 that was present at much higher levels compared with virally delivered Cx32 in this study. It should be noted that direct comparison of treatment outcomes in the three transgenic lines studied here is not possible since they have different genetic backgrounds. Thus, only comparison between treated and mock-treated littermates was performed.
The Cx32 mutants studied here are representative of many CMT1X mutations. The T55I mutant is retained in the ER, while N175D and R75W are retained in the Golgi. It appears that Cx32 mutants retained in the ER are very unlikely to cause any dominant-negative effects on co-expressed WT Cx32. ER retained mutants are misfolded proteins that are less stable and are efficiently eliminated via proteasomal or lysosomal degradation systems (13,19,32). This is also indicated by the lower levels of the T55I mutant compared with the Golgi-retained mutants in our immunoblot analysis (Fig. 1) and in previous studies (21). Along the connexin biosynthesis pathway the formation of hexamers starts in the Golgi (33) so that ER-retained mutant monomers cannot directly interact with WT Cx32. Thus, many other ER-retained Cx32 mutants, including the recently studied A39V and A39P (14), as well as CMT1X mutations occurring in non-coding regions of Cx32 expected to reduce expression levels (7), are predicted to be fully compatible with a gene addition therapy for CMT1X.
In contrast, Golgi retained mutants that exit the ER and are able to reach the Golgi, may have the ability, to different degrees, to participate in hexamer formation. During this process, they could directly interact with WT Cx32 joining the same hexamers, and preventing the correct trafficking of the entire hexamer to the cell membrane. This direct interaction with WT Cx32 has been demonstrated for at least three different Golgi-retained mutants including the R75W and N175D (14) with functional impairment of co-expressed WT Cx32, as previously shown for other mutant connexins like the R142W against WT Cx26 (34). Thus, mutant connexins reaching the Golgi may form mutant-WT hexamers resulting in the disruption of trafficking to the cell membrane where functional GJs should be formed (14,25,35,36). However, it remains unknown whether the Cx32 mutations studied here affect the hexamer formation of the protein and it cannot be predicted by their location since previous studies showed impaired oligomerization of Cx32 caused by mutations in different domains (19). Although the Golgi-retained mutants studied here are representative of several CMT1X mutants retained in the Golgi with or without GJ plaque formation (13,16), not all Golgi retained mutants showed interaction with co-expressed WT Cx32 in vitro (14). Thus, testing of each of them for this possibility, using faster in vitro methods as we previously described (14) would be useful before considering a gene addition therapy in patients harboring a specific mutation.
Based on our results gene addition could be useful only for patients with non-interfering CMT1X mutations such as ER-retained mutants, but not for dominant-negative mutants such as certain Golgi-retained mutants. The latter may be found in up to 50% of all CMT1X families based on the likelihood of Golgi retention according to expression studies published to date and the overall frequency of missense and Inframe INDEL mutations reported (http://hihg.med.miami.edu/code/http/cmt/public_html/index.html#/; date last accessed February 22, 2018). However, we cannot exclude that if high enough expression levels of virally delivered Cx32 are achieved these would be sufficient to overcome the interference of co-expressed mutants. For patients with ER mutations our method could be directly used for the treatment of the disease. For patients with dominant-negative mutations the intrathecal approach could be used alternatively for gene silencing (37,38) followed by gene addition, or gene editing to correct the specific mutations, for example using CRISPR/Cas9 technology.
In conclusion, we show that intrathecal gene delivery of WT Cx32 in different mutant mouse models of CMT1X can improve the functional and morphological properties of demyelinating neuropathy in the presence of mutants that are retained in the ER, but are unlikely to be beneficial for certain mutations that are retained in the Golgi. Further studies will be needed to overcome the limitations of our approach for the dominant-negative mutations in order to develop effective treatments for all CMT1X patients.
Materials and Methods
Cloning and production of lentiviral vectors
Cloning of the lentiviral vectors and vector production and titration methods have been described in our previous studies (26,27). Briefly, for the generation of Schwann-cell targeted lentiviral vectors the expression cassette was cloned into the lentivirus transfer vector pCCLsin.PPT.hPGK.GFP.pre by replacing the PGK promoter with the rat myelin protein zero (Mpz) promoter, which drives strongly gene expression in Schwann cells (28), and by downstream insertion of the human GJB1 open reading frame (ORF), along with the IRES.EGFP. A mock vector was also generated by cloning the Mpz-IRES.EGFP expression cassette without the GJB1 ORF into the same lentiviral vector. Correct assembly of the expression cassettes was confirmed by restriction digest mapping and by sequencing the ORFs.
Recombinant lentiviruses were produced by transient co-transfection of HEK 293T cells with the transfer vector and the helper plasmids (CMVΔR8.74 and pMD2-VSVG for pseudotyping) using the calcium phosphate co-precipitation method. A total of 5× 106 293T cells were seeded in 10 cm plates 24 h prior to transfection in Iscove modified Dulbecco culture medium with 10% fetal bovine serum (FBS), penicillin (100 IU/ml), and streptomycin (100 mg/ml) in a 5% CO2 incubator. One hour prior to transfection the culture medium was changed. A total of 64 μg of plasmid DNA was used for the transfection per dish: 16 μg of the envelope plasmid pMD2-VSVG, 16 μg of the packaging plasmid CMVΔR8.74 and 32 μg of the transfer vector plasmid. The precipitate was formed by adding the plasmids to a final volume of 540 and 60 μl of 2.5 M CaCl2 and then adding drop wise 600 ml of 2× HEPES-buffered saline. The precipitate was added immediately to the cultures. The medium was then replaced after 6 h with fresh medium containing 1 mM Na butyrate. The conditioned medium was collected 60 h after transfection, cleared by low-speed centrifugation, and filtered through 0.22 μm-pore-size filters. The media was concentrated down from 6 ml to 250 μl volume by using the Lenti-X columns (CloneTech). The lentivirus titter was calculated using HIV-1 Gag p24 enzyme-linked immunosorbent assay and one-step RT-qPCR for EGFP.
Experimental animals
All intrathecal gene delivery experiments were conducted using 2-month-old mice expressing different CMT1X mutations on a Cx32 KO background, modeling the genotype of male CMT1X patients expressing a single mutated GJB1 allele. Gjb1-null mice (C57BL/6_129) were originally generated by Prof. Klaus Willecke, University of Bonn, Germany. In these mice, the neor gene was inserted in frame into the Exon 2 of Gjb1 gene which contains the ORF (39). CMT1X models include T55I/Cx32KO (termed T55I KO), which was generated by pronuclear injection of the transgenic cassette described previously (21) on a Bl6/N background (kindly provided by Prof. Charles Abrams, University of Illinois, Chicago, USA), the R75W KO described previously (21) and N175D KO mice (described below), both generated in our transgenic mouse facility on C57BL/6 background, weighing 20–25 g. The T55I mutant has been shown to be retained in the endoplasmic reticulum while the R75W and N175D mutants are retained in the Golgi (13,14). Both male and female mice were used in our experiments and showed no (sex-related) differences in their behavioral performance or nerve pathology. All experimental procedures in this study were conducted in accordance with animal care protocols approved by the Cyprus Government’s Chief Veterinary Officer (project license CY/EXP/PR.L2/2012) according to national law, which is harmonized with EU guidelines (EC Directive 86/609/EEC).
Cell culture and transfections
In order to study the localization of the N175D Cx32 mutant, we transfected HeLa cells with the plasmid DNA carrying the mutation as previously described (14). Communication-incompetent HeLa cells were cultivated in a humidified atmosphere containing 5% CO2 at 37°C in low-glucose Dulbecco’s Modified Eagle’s Medium (DMEM) (GIBCO, Invitrogen, USA) supplemented with 10% FBS, antibiotics–antimycotics (penicillin/streptomycin) at a concentration of 100 μg/ml and 10% non-essential amino acids. Cell cultures were routinely subcloned by trypsinization and change of medium every 2–3 days. Cell transfections were performed using the cationic lipid based Lipofectamine LTX with PLUS reagent transfection system (Invitrogen) when the cells reached ∼70% confluency. For transient transfection, plasmid DNA (WT or mutant) along with the PLUS reagent and lipofectamine LTX were incubated separately in OptiMem (Reduced-Serum Medium) and then mixed for 15 min at RT. The DNA-lipid complex was added dropwise to the cells that were incubated in 4-chamber slides or 6-well plates. The following day, the transfection reagent was replaced with fresh medium.
Generation of N175D transgenic mice
The GJB1/Cx32 N175D ORF was PCR amplified from a construct kindly provided by Dr Donglin Bai (University of Western Ontario). The primers used for the amplification were pIRES-CLA-F 5′-AAATTTACCATCGATGACTCAGATCTCGAGATG-3′ and IRES- R 5′-TCGTGAAGGAAGCAGTTCCT-3′. The digested insert with ClaI and BamHI was directionally cloned into and the pBluescript vector containing the Cnp promoter (gift from Dr Vittorio Gallo, Children’s National Medical Center, Washington, DC, USA) shown to drive expression in myelinating cells (21). Correct assembly of the expression cassette (Supplementary Material, Fig. S1D) was confirmed by restriction digest mapping and sequencing using primers covering the entire coding sequence. Transgenic mice expressing this expression cassette were generated by pronuclear injections. The fragment was isolated, purified and microinjected into the male pronucleus of fertilized oocytes obtained from C57BL/6 mice according to standard protocols. Genotypes of the obtained animals were determined by PCR analysis of the genomic DNA isolated from tail clips as previously described (21). The transgene specific primers were cnp-1F (5′-TGTGGCTT TGCCCATACATA-3′) and Cx32-R (5′-CGCTGTTGCAGCCAGGCTGG-3′) resulting in a 732bp PCR product (94°C ×5 min, 40 cycles of 94°C×30 s, 57°C×30 s, 72°C×30 s and then 72°C×7 min).
Potential founders gave rise to transgenic lines, and each line was screened for the expression of EGFP and Cx32 using immunostaining and immunnoblot analysis. The transgenic lines with best expression for each Cx32 mutation were further expanded for analysis. In addition, in order to generate N175D KO mice on Gjb1-null background, male transgenic mice were bred with female Gjb1-null mice. Genotypes of the offspring were determined using a triple-PCR screening (21) with transgene specific primers already mentioned above, as well as primers for the neorgene (Gjb1- null) (Exon1F: 5′-GACCACTCCCCCTACACAGA-3′; NeoR2: 5′-CTCGTCCTGCAGTTCATTCA-3′) resulting in a 721-bp PCR product (94°C × 5 min, 35 cycles of 94°C × 30 s, 56°C × 30 s, 72°C × 30 s and then 72°C × 7 min); and primers specific for the wild-type (WT) Gjb1 mouse gene (Exon1F and Cx32R).
Intrathecal vector delivery
We delivered the lentiviral vector as previously described (26). Briefly, following a small skin incision along the lower lumbar spine level to visualize the spine, the lentiviral vector was delivered into the L5–L6 intervertebral space of anesthetized mice. A 50-μL Hamilton syringe (Hamilton, Giarmata, Romania) connected to a 26-gauge needle was used to inject 30 μL of lentiviral stock containing an estimated 4.0× 1011 to 5.3×1012 (full vector) and 7.5×1011 to 9.9×1012 (mock vector) viral particles/ml. A flick of the tail was considered indicative of successful intrathecal administration.
Immunoblot analysis
Fresh sciatic nerves were collected from 2-month-old mice from the three different mutant lines on Cx32 KO background and lysed in ice-cold RIPA buffer (10 mM sodium phosphate, pH 7.0, 150 mM NaCl, 2 mM EDTA, 50 mM sodium fluoride, 1% Nonidet P-40, 1% sodium deoxycholate, and 0.1% SDS) containing a mixture of protease inhibitors (Roche, Mannheim, Germany). Proteins (150 μg) from the lysates were fractionated by 12% SDS/PAGE and then transferred to a Hybond-C Extra membrane (GE Healthcare Bio-Sciences, USA) using a semidry transfer unit. Nonspecific sites on the membrane were blocked with 5% non-fat milk in PBS with Tween 20 (PBST) for 1 h at room temperature (RT). Immunoblots were incubated with rabbit antisera against EGFP (1:1000; Abcam, Cambridge, UK) or Cx32 [clone 918, 1:3000 (40)] or with mouse anti-Cx32 (1:1000; Invitrogen) and for loading control with β-tubulin antibody (1:4000; Developmental Studies Hybridoma Bank, USA) at 4°C overnight. After washing, the immunoblots were incubated with appropriate anti-mouse or anti-rabbit HRP-conjugated secondary antisera (1:3000; Jackson ImmunoResearch, USA) in 5% milk–PBST for 1 h. The bound antibody was visualized by an enhanced chemiluminescence system (GE Healthcare Bio-Sciences, USA). The intensity of the bands was measured with Tinascan version 2.07d and normalized for loading with the β tubulin bands.
Immunohistochemistry
For immunostaining, mice were anesthetized with avertin according to institutionally approved protocols, and then transcardially perfused with normal saline followed by fresh 4% paraformaldehyde in 0.1 M PB buffer. The lumbar-sacral spinal cord with spinal roots attached as well as the bilateral sciatic and femoral motor nerves were dissected. All tissues were frozen for cryosections, while sciatic and femoral nerves were isolated and teased into fibers under a stereoscope. Teased fibers or sections were permeabilized in cold acetone and incubated at RT with a blocking solution of 5% BSA (Sigma-Aldrich, Munich, Germany) containing 0.5% Triton-X (Sigma-Aldrich) for 1 h. Primary antibodies used were mouse monoclonal antibody against contactin-associated protein (Caspr, 1:50; gift of Dr Elior Peles, Weizmann Institute of Science), 58k (1:50; Abcam, Cambridge, UK), Kv1.1 (1:500; Upstate Biotechnology, USA), GFAP (1:400; Sigma-Aldrich), NeuN (1:400; Chemicon), CC1 (1:50; Calbiochem, USA); rabbit antisera against EGFP (1:1000; Invitrogen, USA), Cx32 (1:50; Sigma, Munich, Germany) and Iba1 (1:500; Biocare, USA), all diluted in blocking solution and incubated overnight at 4°C. Slides were then washed in PBS and incubated with fluorescein- and rhodamine-conjugated donkey, goat, and rabbit cross-affinity purified secondary antibodies (1:500; Jackson ImmunoResearch, USA) for 1 h at RT. Cell nuclei were visualized with DAPI (1 µg/ml; Sigma). Slides were mounted with fluorescent mounting medium and images photographed under a fluorescence microscope with a digital camera using Axiovision software (Carl Zeiss MicroImaging; Oberkochen, Germany).
Gene therapy study design
The aim of this study was to examine whether a gene addition method can prevent the development of peripheral neuropathy in CMT1X mutant mouse models. To perform this treatment trial we used 2-month-old mice from three different CMT1X mutant lines, T55I KO, R75W KO and N175D KO mice. T551 mutants were generated by pronuclear injection of the transgenic cassette described in (21) on a Bl6/N background and then crossed with the backcrossed Cx32 KO line. R75W and N175D were generated by pronuclear injection of the transgenic cassette described in (21) on a C57BL/6 background and then crossed with the backcrossed Cx32 KO line. The gene therapy trial was conducted using two groups of each of the CMT1X mutant lines on Cx32 KO background. A minimum of 8–12 mice per treatment group for each outcome measure was considered adequate for assessing statistically significant differences based on our previous studies using similar models (26,27). Animals were treated by a single intrathecal injection (above) at the age of 2 months, before nerve pathology develops (21). Littermate mice were randomized to receive either LV.Mpz-GJB1 (full vector) treatment or LV.Mpz-Egfp (mock vector treatment, as a control group) and assigned a coding number for further identification. Mice were evaluated by motor behavioral testing (methods below) at the age 4 and 8 months and sacrificed at the age of 8 months for quantitative morphometric analysis (methods below) of semithin sections (n = at least 8 animals per treatment group). Both the behavioral as well as the morphometric analyses were performed by examiners who were blinded to the treatment condition.
Behavioral analysis
Rotarod testing
Motor balance and coordination was determined as described previously (41) using an accelerating rotarod apparatus (Ugo Basile, Varese, Italy). Training of animals consisted of three trials per day with 15-min rest period between trials, for 3 consecutive days. The mice were placed on the rod and the speed was gradually increased from 4 to 40 rotations per minute (RPM). The trial lasted until the mouse fell from the rod or after the mouse remained on the rod for 600 s and was then removed. Testing was performed on the fourth day using two different speeds, 20 and 32 rpm. Latency to fall was calculated for each speed. Each mouse was placed on the rotarod three times at each speed used and three different values were obtained for each speed. Mean values were used for each mouse at the two different speeds.
Grip strength testing
To measure grip strength, mice were held by the tail and lowered toward the apparatus (Ugo Basile, Varese, Italy) until they grabbed the grid first with all four paws and then with only the front paws. Mice were gently pulled back until they released the grid. Measurements of the force in gravity were indicated on the equipment. Each session consisted of three consecutive trials and measurements were averaged. Hind limb force was compared between LV.Mpz-GJB1 and LV.Mpz-Egfp treated mice.
Morphometric analysis of lumbar roots and peripheral nerves
Mice were transcardially perfused with 2.5% glutaraldehyde in 0.1 M PB buffer. The lumbar spinal cord with multiple spinal roots attached, as well as the femoral and sciatic nerves, were dissected and fixed overnight at 4°C, then osmicated, dehydrated, and embedded in araldite resin (all purchased from Agar Scientific, Essex, UK). Transverse semithin sections (1 μm) of the lumbar spinal cord with roots and the middle portion of the femoral motor and sciatic nerves were obtained and stained with alkaline toluidine blue (Sigma-Aldrich). Sections were visualized with 10×, 20×, and 40× objective lenses and captured with a Zeiss AxioCam HR camera. Images of whole root or transverse nerve sections were obtained at 100–200× final magnification, and a series of partially overlapping fields covering the cross-sectional area of the roots or the nerves were captured at 400× final magnification. These images were used to examine the degree of abnormal myelination in both groups as described previously (21,27,42). In brief, all demyelinated, remyelinated, and normally myelinated axons were counted using the following criteria: axons larger than 1 μm without a myelin sheath were considered demyelinated, axons with myelin sheaths <10% of the axonal diameter and/or axons surrounded by ‘onion bulbs’ (i.e. circumferentially arranged Schwann cell processes and extracellular matrix) were considered remyelinated, and other myelinated axons were considered normally myelinated.
In addition, we counted the number of foamy macrophages present within the entire cross section of each root or nerve, as an indication of inflammation. Macrophages were identified in semithin sections at 400× magnification as cells laden with myelin debris, devoid of a basement membrane, and extending small, microvilli-like processes, as described previously (43,44). The macrophage count was calculated as the ratio per 1000 myelinated fibers, to account for size differences between different spinal roots and nerves. All pathological analyses were performed blinded to the treatment condition of each mouse.
Statistical analysis
The percentages of Cx32-expressing paranodal myelin areas in immunostained spinal roots and sciatic nerves of different mutant lines injected with the full vector were compared with Student’s t-test. Behavioral testing results as well as morphological analysis data including the proportion of abnormally myelinated fibers and the number of macrophages in lumbar spinal roots and peripheral nerves obtained from mock-treated and fully treated mice during the treatment trial were compared using the unpaired two-sided Student’s t-test as well as the Mann–Whitney U test (significance level for all comparisons, P < 0.05) using Graphad Instat3 software (GraphPad, USA).
Supplementary Material
Supplementary Material is available at HMG online.
Acknowledgements
We thank Dr Donglin Bai (University of Western Ontario) for providing the N175D construct, Dr Charles Abrams (University of Illinois, Chicago) for providing the T55I mice, Dr Vittorio Gallo (Children’s National Hospital, Washington, DC) for the Cnp promoter construct, Dr Elior Peles (Weizmann Institute of Science) for the Caspr antibody, and Dr Steven Scherer (University of Pennsylvania) for valuable comments. We also thank Drs Natasa Schiza and Styliana Kyriakoudi for technical assistance and Dr Kyriacos Kyriacou for advice on processing semithin sections. The monoclonal antibody E7 against β-tubulin (developed by Michael Klymkowsky) was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the National Institute of Child Health and Human Development and maintained by Department of Biology, The University of Iowa.
Conflict of Interest statement. None declared.
Funding
This work was supported by a Muscular Dystrophy Association (MDA) Grant (MDA277250, to K.A.K.) and by an MDA/Charcot-Marie-Tooth Association (CMTA) Grant (MDA480030 to K.A.K.). Funding to pay the Open Access publication charges for this article was provided by the Cyprus Institute of Neurology and Genetics.
