Abstract

   The early life environment significantly affects the development of
   age-related skeletal muscle disorders. However, the long-term effects
   of lactational protein restriction on skeletal muscle are still poorly
   defined. Our study revealed that male mice nursed by dams fed a
   low-protein diet during lactation exhibited skeletal muscle growth
   restriction. This was associated with a dysregulation in the expression
   levels of genes related to the ribosome, mitochondria and skeletal
   muscle development. We reported that lifelong protein restriction
   accelerated loss of type-IIa muscle fibres and reduced muscle fibre
   size by impairing mitochondrial homeostasis and proteostasis at 18
   months of age. However, feeding a normal-protein diet following
   lactational protein restriction prevented accelerated fibre loss and
   fibre size reduction in later life. These findings provide novel
   insight into the mechanisms by which lactational protein restriction
   hinders skeletal muscle growth and includes evidence that lifelong
   dietary protein restriction accelerated skeletal muscle loss in later
   life.

   Keywords: Skeletal muscle, Sarcopenia, Maternal nutrition, Protein
   restriction, Proteostasis, Mitochondrial homeostasis

Graphical abstract

   [39]Image 1
   [40]Open in a new tab

Highlights

     * •
       Lifelong protein restriction accelerates muscle fibre loss and
       reduces fibre size.
     * •
       Lifelong protein restriction disrupts mitochondrial homeostasis in
       skeletal muscle.
     * •
       Lifelong protein restriction may impair proteostasis in skeletal
       muscle in mice.
     * •
       The detrimental effects of lactational protein restriction can be
       corrected.
     * •
       The adverse effects of suboptimal early life depend on its timing.

1. Introduction

   The nutritional environment experienced in early life can influence the
   risk of developing age-related diseases, including type 2 diabetes,
   cardiovascular diseases, and sarcopenia [[41][1], [42][2], [43][3],
   [44][4]]. Maternal protein malnutrition is a concern in many
   populations globally, and it leads to permanent structural changes in
   organs, alterations in gene expression, and changes in cellular ageing
   in the offspring [[45]1,[46][5], [47][6], [48][7]]. It is widely
   accepted that maternal protein restriction during gestation limits the
   in utero growth of the offspring and has long-lasting effects that
   persist into later life through epigenetic modifications
   [[49]1,[50]8,[51]9]. However, it is unclear whether maternal protein
   restriction during lactation protects the offspring through hormesis or
   has similar effects to maternal protein restriction during gestation
   [[52]10,[53]11]. Reports from animal studies have demonstrated that
   lactational protein restriction impairs major metabolic pathways
   involved in the regulation of lifespan and metabolic syndrome in
   21-day-old mice or in adult rats [[54][12], [55][13], [56][14]], but
   pups from dams on a low protein diet demonstrated an increased lifespan
   compared to control mice [[57]15].

   Muscle growth occurs through an increase in muscle fibre number or an
   increase in the size of individual fibres [[58]16]. While postmitotic
   muscle fibres are formed during embryonic development via primary and
   secondary myogenesis [[59]17], the early neonatal period in humans and
   rodents is also a critical time for skeletal muscle development as
   muscle hypertrophy mainly occurs during early postnatal life through a
   rapid increase in the number of myonuclei [[60]18,[61]19]. Skeletal
   muscle is a highly dynamic tissue, and its development is particularly
   prone to nutritional deficiency compared with other tissues [[62][20],
   [63][21], [64][22]]. Suboptimal maternal nutrition during development
   dysregulated myogenesis, reduced the number of muscle fibres and muscle
   mass, and changed fibre type distribution in the offspring
   [[65]22,[66]23]. Moreover, it has been shown that an isocaloric
   maternal low protein diet impacted skeletal muscle metabolism and mass,
   and the expression of key genes involved in mitochondrial metabolism in
   skeletal muscle of the offspring [[67]12,[68][24], [69][25], [70][26]].
   Although lactational protein restriction reduced skeletal muscle weight
   in mice at weaning, it remains unclear whether these adverse effects
   persist in skeletal muscle following weaning onto a normal protein diet
   in later life [[71]12].

   In this study, we investigated the effects of lactational protein
   restriction on global gene expression changes in skeletal muscle at
   weaning and identified that genes involved in ribosomal homeostasis,
   mitochondrial function, myofibre and muscle development were affected.
   We further demonstrated that lifelong feeding a low-protein diet after
   lactational protein restriction accelerated type-IIa fibre loss and
   reduced muscle fibre size at 18 months of age by dysregulating
   ribosomal gene expression, autophagy, AMP-activated protein kinase
   (AMPK) signalling, and mitochondrial homeostasis. Some of the adverse
   effects of maternal protein restriction during lactation on skeletal
   muscle were corrected following weaning onto a normal protein diet in
   18-month-old male mice.

2. Materials and methods

2.1. Animals

   B6. Cg-Tg (Thy1-YFP)16Jrs/J mice (Jackson Laboratory; stock number
   003709), were used for this study. Individual vented cages were
   utilized to house the mice with a fixed light cycle (21 ± 2 °C, 12-h
   light/dark cycle). Ethical approval was obtained from the University of
   Liverpool Animal Welfare Ethical Review Committee (AWERB), and UK
   Animals (Scientific Procedures) Act 1986 regulations were followed for
   all experimental protocols for the handling and use of laboratory
   animals.

2.2. Experimental design

   Two weeks prior to mating, nulliparous female mice were fed either a
   low-protein diet (L, 8 % crude protein; Special Diet Services, UK; code
   824,248) or a normal-protein diet (N, 20 % crude protein; Special Diet
   Services, UK; code 824,226), both fed ad libitum. Age-matched male mice
   on a normal-protein diet were used for mating. When the female mice
   were pregnant, they were kept on the same diet. The male offspring from
   mothers on normal-protein diet during gestation were used for this
   study. Within 24 h of birth, the new-born male litters were
   cross-fostered to mothers fed either a normal-protein diet or a
   low-protein diet, in order to create Normal-Normal (NN) or Normal-Low
   (NL) groups of mice, respectively. Suckling pup numbers were kept the
   same for all animals during lactation (n = 6 pups). At the end of
   weaning period (day 21), some NN and NL mice were humanly culled for
   RNA sequencing (RNA-seq) (n = 6). Other NN mice were fed a normal
   protein diet generating Normal-Normal-Normal (NNN or control, n = 6)
   mice ([72]Fig. 1a). In order to understand the effects of lactational
   protein restriction and to establish if this could be corrected by
   feeding a normal diet from weaning, rest of NL mice were weaned onto
   either a normal protein diet, creating Normal-Low-Normal (NLN, n = 6),
   ([73]Fig. 1a). Moreover, NL mice were fed a low-protein diet to
   investigate the effects of lifelong protein restriction, generation
   Normal-Low-Low (NLL, n = 6), ([74]Fig. 1a). Mice were fed ad libitum
   food and water. At 3-month-old or 18-month-old, mice were sacrificed.
   Body weights were recorded immediately, prior to dissection.
   Gastrocnemius (GAS) and soleus (SOL) muscles were carefully dissected
   and weighed. Muscle tissues were stored at −80 °C until analysis and
   prepared for the experiments as described below.

Fig. 1.

   [75]Fig. 1
   [76]Open in a new tab

   Illustrative description of experimental design, body weight and muscle
   weight. (a) Normal-Normal-Normal (NNN) litters received a normal
   protein diet during gestation and postnatally. Normal-Low (NL)-Normal
   (NLN) litters nursed by low-protein-fed dams during lactation and
   received a normal protein diet after weaning. Normal-Low (NL)-Low (NLL)
   litters were nursed by low-protein-fed dams during lactation and kept
   on a low-protein diet after weaning. (b) Body weights (g) of NNN, NLN,
   and NNN mice at 21-day-old, 3-month-old, and 18-month-old. n = 6. (c)
   GAS muscle weights (g) of NN and NL mice at 21-day-old. n = 6. (d, e)
   GAS muscle weights (g) of NNN, NLN, and NLL mice at (d) 3-month-old,
   (e) 18-month-old of age. n = 6. (f, g) SOL muscle weights (g) of NNN,
   NLN, and NLL mice at (d) 3-month-old, (f) 18-month-old of age. n = 6.
   Results are expressed as the mean ± standard deviation (mean ± SD). *
   ++++ p < 0.0001 shows the significant difference between NN and NL mice
   in [77]Fig. 1b. *p < 0.05, * *p < 0.01, * * *p < 0.001, *** * *
   *p < 0.0001 shows the significant difference between NNN and NLL mice.
   Statistical comparisons were performed using ordinary one-way ANOVA
   with a Dunnett's multiple comparisons test, considering NNN as the
   control group.

2.3. RNA isolation, library preparation and sequencing

   Frozen GAS skeletal muscles of 21-day-old mice (n = 6) were ground
   using a mortar and pestle in liquid nitrogen. Total RNA from frozen GAS
   muscle was extracted and purified by RNeasy mini kits with on-column
   DNase treatment (Qiagen, Manchester, UK) following manufacturer's
   instructions. Preparation of dual-indexed, strand-specific RNA-seq
   library from submitted total RNA was performed using the NEBNext polyA
   selection and Ultra II Directional RNA library preparation kits (New
   England Biolabs, UK) as described previously [[78]27]. Total RNA
   integrity was confirmed by an Agilent 2100 Bioanalyzer (Agilent
   Technologies, Santa Clara, CA, USA) with RNA Integrity Number
   (RIN) > 7. RNA-seq carried out by Illumina NovaSeq using S1 chemistry
   (paired-end, 2 × 150bp sequencing, generating an estimated 650 million
   clusters per lane). Number of reads obtained by RNA-seq were provided
   in [79]Supplementary Fig. 4. Sequencing was performed at the Centre for
   Genomic Research, University of Liverpool
   ([80]https://www.liverpool.ac.uk/genomic-research/). The raw RNA-seq
   data have been deposited in NCBI Gene Expression Omnibus (GEO) database
   under accession number [81]GSE235750
   .

2.4. Data processing and bioinformatic analyses

   The raw Fastq files are trimmed by the Centre for Genomic Research,
   University of Liverpool for the presence of Illumina adapter sequences
   using Cutadapt version 1.2.1 [[82]28]. Any reads which match the
   adapter sequence for 3 bp or more are trimmed. The reads are further
   trimmed using Sickle version 1.200 with a minimum window quality score
   of 20. Reads shorter than 15 bp after trimming were removed. The
   processed reads were then pseudo-aligned to the Ensemble release mus
   musculus transcriptomes v96 with Kallisto quant version 0.46.1 to get
   the transcripts length and abundance estimates [[83]29] The output from
   these tools was converted count by tximport (TXI) version 1.26.1. in R
   version 4.2.2 [[84]30]. An average of 41.96 million total counts for NN
   and 45.32 million total counts for NL were generated following
   alignment ([85]Fig. 2a). Differentially expression and statistical
   analyses were carried out by DESeq2 version 1.38.3. and PCA plots were
   visualised by plotPCA function in ggplot2 version 3.4.1 in R. We used
   the following cutoffs:|Log2 Fold Change| (|Log2FC|)>0.3 and adjusted
   p-values<0.1 to compare gene expressions. To generate the heatmap, we
   selected the 130 genes which had |Log2FC|>1 and adjusted p-values<0.1
   using the ‘pheatmap' library version 1.0.12 in R.

Fig. 2.

   [86]Fig. 2
   [87]Open in a new tab

   Identification of genes associated with the growth restriction
   phenotype seen in 21-day-old lactational protein-restricted mice. (a)
   Total count number of NN and NL samples. n = 6. (b) Gene count of
   differentially expressed genes in skeletal muscle from NL vs NN mice.
   (c) Principal component analysis indicated NN mice were grouped
   together, whilst NL mice exhibited a spread pattern along two principal
   components PC1 (37 % of the total variance) and PC2 (13 % of the total
   variance). n = 6. (d) Volcano plot summarising upregulated (green) and
   downregulated (blue) genes |Log2FC|>1 and adjusted p-values<0.1 in NL
   mice. Genes with adjusted p-values<0.1 and |Log2FC|<1 are highlighted
   on the middle of the plot and illustrated with red dots. (e) Heatmap
   summarising expression levels of upregulated or downregulated genes
   (|Log2FC|>1 and adjusted p-values<0.1) in groups. The gene expression
   level is presented as the read count. The colour code scale indicates
   the normalized counts (ranging from −3 (red) up to 3 (blue)). (For
   interpretation of the references to colour in this figure legend, the