Morphological analysis of hair cycle changes in goat skin
We made a histological examination of skin tissue from cashmere goats, as follows. Results showed (Fig. 1), that the number of secondary hair follicles in cashmere goats decreased gradually from December to March (Fig. 1L, A, B, C). The lowest value was reached in March (Fig. 1C), and the statistical values of each trait were also lower. The division and extension of hair follicles to the dermis began in April (Fig. 1D), while the number of secondary hair follicles also began to increase at the same time. A velvet-like appearance to the goat’s coats was observed in the month of July (Fig. 1G). Most cashmere grew from follicles in the skin between August and September (Fig. 1H, I). At the same time, the number of secondary follicles reached its highest level, with this period being considered as the peak period of cashmere growth. In October, hair follicle bulb cells began to enlarge, gradually aged and died, and the dermal papillae began to atrophy. The numbers of secondary hair follicles gradually decreased (Fig. 1J). In December, the hair follicle roots rose to the sebaceous glands, and the secondary follicle numbers reached their lowest level (Fig. 1L), This state was maintained until February of the following year. Fro this information, we made the initial inference that the cycle of secondary hair follicles in cashmere goats can be divided into a growth period from March to September, a resting period from September to December, and a regression period from December to March. Generally speaking, we divided the cashmere hair cycle into three periods by observing skin tissue morphology, but the key points of each time period could not be determined; this needs further study.
Differential gene expression analysis
We first screened and analyzed data quality (Table 1) and data length distribution (Table 2). Then we compared the skin transcriptome data of cashmere goats in 12 months with neighboring months(Fig. 2). It was found that the number of differentially expressed genes was the greatest between February and March. The total number of differential genes was 1059, of which 219 were up-regulated and 840 were down regulated. In March and April, the number of differentially expressed genes was 731, of which 550 were up-regulated and 181 were down regulated. In June and July, there were 418 differentially expressed genes, of which 388 were up-regulated and 30 were down regulated. These results showed that the expression of genes was initially up-regulated or down-regulated during the initiation of secondary hair follicle growth. Along with advancement in hair follicle initiation, the number of down-regulated genes began to decrease, and the number of up-regulated genes continued to increase. After completion of the initiation process, the gene changes tended to be stable. The results further showed that hair follicle development was initiated by a combination of up-regulation and down-regulation of genes in the early stage of initiation, and that gene expression returned to normal levels after initiation. Comparing the data between June and July, we found that there was another significant change in gene expression during cashmere outgrowth. We believe that this change promoted cashmere to emerge from the skin surface, but the existence of other roles remains to be studied. From August to February of the following year, secondary hair follicle gene expression changed significantly from quiescence to degeneration. These results further showed that the initiation of secondary hair follicles in cashmere goats began in March. Finally, it is worth mentioning that the number of differentially expressed genes increased first and then decreased from February to March, and then to April, thus emphasizing that the secondary follicle cycle starts in March.
Classification of gene function annotation
According to GO classification statistics, skin expression genes can be divided into three main categories: biological functions, cell components, and molecular functions. In this study, 51 078 transcripts were noted using GO annotation. Among them, in biological function, the most annotated transcript was cellular process. In cellular component, most of the transcripts were transcribed to cell and cell part. In molecular function, most of the transcripts were transcribed to binding. It is speculated that during the hair follicle cycle, the changes of gene expression led to changes in the number and state of cells in hair follicles, which further led to the occurrence or shedding of secondary hair follicle.
Group clustering analysis of natural periodic samples
In order to further explore the rule of gene expression, we calculated the correlation coefficient and cluster analysis of all gene expression levels in the 12-month natural cycle (Fig. 4). The results showed that clustering information could be divided into three categories. The sample LZH3 was isolated because of the great changes in gene expression of the follicle promoter. Samples LZH2–LZH7 were considered to be the initiation process of hair follicle growth. Samples LZH8–LZH12 were clustered together because they were thought to control the transition of secondary hair follicles from vigorous growth to recession. Gene expression remained relatively unchanged from December to March, so the clustered sites had reached the end of degeneration and before growth; this was considered to be the resting period of hair follicle development. Combined with previous studies in this manuscript, we found that there were several critical periods in the division of the secondary hair follicle cycle. March is considered the key point for initiation of the hair follicle cycle; September is the key period for vigorous growth and the beginning of recession; December is the key point for the end of hair follicle recession and the beginning of the rest period, These three critical periods were determined by key signals in hair follicle and cashmere growth.
Extraction and analysis of target gene expression information
In order to explore the expression patterns of genes that play key roles in the cycle, we extracted expression information of all target genes for 1–12 months, and then clustered the expression patterns by analysis and exclusion. Gene expression patterns of several pathways related to the cashmere cycle were obtained (Fig. 5A). Results showed that gene expression patterns related to the cashmere cycle were consistent with our analysis of differential gene expression. The results further supported the previous finding that the hair follicle cycle was initiated in March, entered the regression stage in September, and entered the end stage in December. However, transition through the hair follicle cycle cannot be visualized through skin histology, because the cycle initiation precedes cashmere growth, and there is a causal relationship between them. The period of cashmere growth was observed in tissue sections, and there was a direct relationship between cashmere growth and the expression of keratin. Therefore, in order to further verify the cashmere growth cycle, we also clustered the expression patterns of keratin and keratin-related genes (Fig. 5B). The results showed that the expression of keratin was consistent with the results of tissue sections, which further supported our findings that the hair follicle cycle first started (degenerated or rested), and then cascaded, leading to changes in the expression of the keratin gene, thus promoting the occurrence of cashmere (growth or degeneration). The gene expression characteristics can be divided into two types. First, the expression of apoptosis-related genes showed a downward trend from the resting stage to the early growth stage (Figure 5B, LZH3). Secondly, the expression of genes increased which related to hair follicle development (Fig. 5A, LZH3) or decreased after the development of hair follicles during the growth period (Fig. 5B, LZH6), and the expression of genes related to controlling cashmere growth increased (Fig. 5A, LZH6). In general, development of hairfollicles and cashmere growth showed a wave-like expression. In order to further study the relationship between the keratin gene and cycle, we selected the gene with the highest expression as a case for further study.
QPCR analysis of key genes
Among keratin and keratin associated proteins, keratin associated protein 3-1 ranked first in cluster 11 with an expression level of 80 824 at the growth stage and 23 856 at rest stage. The expression of keratin associated protein 3-1 in cashmere for 12 months was confirmed by quantitative PCR (Fig. 6A). The results showed that KAP 3-1 was expressed in the skin at different stages of the year, its expression was significantly different (P < 0.05) and fluctuated periodically in a 12 month period. Expression levels in the three months of August, September, and October were significantly higher than that in other months (P < 0.05). Combined with previous studies, the expression quantity was verified by investigating different periods (Fig. 6B). It was found that expression of the KAP3-1 gene in the growth phase was significantly higher than that in either the rest or regression phases. Subsequently, to verify the stability of gene expression, we examined the expression of two other genes, KAP 8-1 and KAP 24-1, in cashmere goat skin at several stages using fluorescence quantitative analysis (Fig. 7). The relative expression of KRTAP 8-1 (Fig. 7A) and KRTAP 24-1 (Fig. 7B) genes in Inner Mongolian cashmere goat skin showed periodic variation, which was consistent with the hair follicle development cycle. This indicated that KRTAP 8-1 and KRTAP 24-1 genes played a positive role in controlling cashmere wool growth and were closely related to the regulation of cashmere growth and cycle transformation.