Southern blot analysis for transgene copy number. Southern blot analysis (B) revealed ten founders (F80), from which three lines were established (Tg80). A polymorphic BglII restriction site (B*) distinguishes the mP₀TOT transgene from the endogenous Mpz FVB/N alleles (A). Copy numbers (CN) were estimated by the ratio of the 2.8–6.8-kb signal intensities.

Semi-thin section analysis of sciatic nerve in Tg80.2. Transverse (A and C) or longitudinal (B and D) sections stained with toluidine blue from P28 Tg80.2 (C and D) as compared with wild-type (A and B) sciatic nerves showed a marked paucity of myelin (one internode of thin myelin is marked by a double asterisk) and an increased number of nuclei (arrowheads). In addition, bundles of unsorted axons surrounded by several Schwann cell nuclei were evident in every section (arrows; see also Fig. 4). No onion bulbs were seen. Bar, 20 μm.

Semi-thin section analysis of Tg80.2 sciatic nerves during postnatal development. Transverse sections stained with toluidine blue from wild-type (A, C, E, and G) or Tg80.2 (B, D, F, and H) animals at P5 (A and B), P15 (C and D), P45 (E and F), and P90 (G and H) demonstrate that dysmyelination was already present by P5 and remained throughout development. Arrows indicate bundles of unsorted axons. Extracellular myelin debris was not obvious at any age (see also Fig. 4). Bar, 50 μm.

Semiquantitative RT-PCR analysis reveals Mpz overexpression. A, Total RNA prepared from sciatic nerve was reverse-transcribed; the product was amplified by PCR using a primer pair that recognized identically both endogenous (FVB/N) or Tg80 (BALB/c) P₀ cDNAs. The FVB/N cDNA was distinguished by DpnII digestion, revealing 244- and 82-nucleotide (nt) products. B, Various mixtures of FVB/N (F) and BALB/c (B) reverse-transcribed product were analyzed to validate the method. The relative proportion (in percent of total signal) of FVB/N (326-nt band) or BALB/c (244 + 82-nt bands) were ± 5% of predicted in three separate experiments (one shown). C, Reverse-transcribed products from sciatic nerve total RNA from various transgenic nerves at P28 or P2 (Tg80.2 P2) were analyzed for the relative proportion of Tg80 P₀ mRNA. Overexpression (-fold) was the quotient of Tg80 signal divided by endogenous signal. Note that the pairs, Tg80.4 and Tg80.3H (homozygote) or Tg80.2 P28 and Tg80.2 P2, manifested very similar ratios of overexpression. Undigested amplification product (uncut) revealed only the 326-nt band.

P₀ overexpression is accompanied by alteration of other myelin protein levels. Western blot analysis was performed on P28 sciatic nerve lysates from Tg80 animals with dysmyelination ranging from none (Tg80.3), to moderate (Tg80.4), to severe (Tg80.2). Dysmyelination was associated with proportionally decreased total P₀, and high molecular weight isoforms of MBP consistent with developmental delay. In contrast, Tg80.3 nerves showed ∼60% increased P₀ and mature MBP isoforms. Note that MBP and PMP22 levels were reduced in Tg80.3 nerves, even when normalized to tubulin or amido black staining of the blot for total protein. Numbers × 1,000 indicate approximate M_r.

Phenotype of high copy number Tg80.2 animals. A, Tg80.2 animals at P120 manifested atrophy in axial and hindlimb muscles. B, Traces are shown of CMAP generated by sciatic nerve stimulation in P42 animals. Arrows indicate stimulation artifact. Note the markedly delayed onset and dispersion of the Tg80.2 CMAP, due to the slowed NCV. C, ATPase isotype staining (pH 4.6) in leg muscles revealed atrophy of many fibers and angulated fibers (arrows), suggesting denervation. Bar, 100 μm.

Ultrastructural analysis of Tg80.2 sciatic nerve showed dysmyelination. A, A typical transverse section of P28 sciatic nerve showed a mixture of occasional Schwann cells that had formed thin myelin sheaths of normal periodicity (inset), or more commonly, Schwann cells that had ensheathed single large axons, but had not advanced an inner mesaxon around them (asterisks). Nonmyelin-forming Schwann cells (nmsc) ensheathed small axons normally. Bar, 5 μm. B, Other Schwann cell families surrounded bundles of mixed caliber, naked axons; some of these Schwann cells were segregating large axons away from the bundles (arrow). Bar, 2 μm. C, Many Schwann cells were surrounded by pockets of redundant basal lamina, some containing collagen fibers (arrowheads). Bar, 1 μm.

Dysmyelination parallels Mpz gene dosage. Low magnification electron micrographs of transverse sections of P28 sciatic nerve from FVB/N (A), Tg80.3 (B), Tg80.4 (C), Tg80.3 homozygote (D), Tg80.2 (E), and F80.8 (F) demonstrate that dysmyelination was absent in low copy number animals (B), moderate in middle copy number animals (C), and most pronounced in high copy number animals (E and F). Note that in contrast to Tg80.3 hemizygotes (B), Tg80.3 homozygotes manifested dysmyelination (D). Bar, 5 μm.

Dysmyelination is rescued in the Mpz null background. Tg80.4 was crossed into the Mpz heterozygous and homozygous null background and offspring were identified by Southern blotting (G). The Mpz null allele produced a 2.8-kb band due to the BglII polymorphism (see Fig. 1), and an additional 2.1-kb band due to the neomycin resistance gene. In the homozygous null animals, no 6.8-kb band appeared as the endogenous FVB/N Mpz alleles were absent. Tg80.4 was distinguished (both the Mpz null allele and Tg80.4 contain the BglII polymorphism) by the presence of a much more intense 1.0-kb band. Semi-thin section analysis of P28 sciatic nerves from 3–5 animals of each genotype showed that Tg80.4 homozygous null nerves (80.4/^−/−; D) appeared normally myelinated as compared with nontransgenic (FVB/N; A) littermates, and as compared with the dysmyelinated Tg80.4 (B) or homozygous null (Mpz^−/−; C) littermates. Note that at P28, heterozygous null (Mpz^+/−; E) littermates were slightly hypomyelinated in the FVB/N congenic background, and that Tg80.4 heterozygous null nerves (80.4/^+/−; F) showed a mixture of hypo- and hypermyelinated fibers. Bar, 20 μm.