This is an interesting question! I had to do some quick reading to verify the answer as I haven’t considered this sort of stuff for many many years. I was very surprised to find out that there are considerably more lineage steps in sperm production than I remembered!
Here is a diagrammatic representation of the cell lineages that give rise to mature sperm......
<img src="http://www.ncbi.nlm.nih.gov/books/bookres.fcgi/dbio/ch19f18.jpg">
What I found out is this:
Gene transcription occurs through all the spermatogonial stages. This isn’t surprising as these are “stem cell” types. What did surprise me was that even the haploid spermatocyte cells are able to synthesize certain products. Much of gene transcription during spermatogenesis (ie. the meiotic division of spermatocytes that give rise to spermatids) takes place during the meiotic prophase. The genes that are transcribed specifically during spermatogenesis are often those whose products are necessary for sperm motility or binding to the egg.
Examples: RNA-binding proteins are essential for spermatogenesis because many of the genes expressed in the sperm are regulated at the level of translation. A sperm specific isoform of beta-tubulin is seen only during spermatogenesis, and it is responsible for forming the meiotic spindles.
The haploid cells thus formed are called spermatids. In addition to gene transcription in diploid cells during meiotic prophase, certain genes are transcribed in the spermatids. Now, spermatids are still connected to one another through their cytoplasmic bridges. The spermatids that are connected in this manner have haploid nuclei
but are functionally diploid, since a gene product made in one cell can readily diffuse into the cytoplasm of its neighbors. Wow!
The mammalian haploid spermatid is a round, unflagellated cell that looks nothing like the mature vertebrate sperm. The gene for the beta-1,4-galactosyltransferase that binds the sperm to the zona pellucida is transcribed only during the haploid phase of sperm maturation.
The next step in sperm maturation, then, is spermiogenesis (or spermateliosis), the differentiation of the sperm cell. Transcription of the gene for protamine is seen in the early haploid cells (round spermatids), although their translation is delayed several days. One of the major changes in the nucleus is the replacement of the histones by protamines. During spermiogenesis, the nucleosomes dissociate, and the histones of the haploid nucleus are eventually replaced by protamines. This causes the complete shutdown of transcription in the nucleus and facilitates its assuming an almost crystalline structure. The resulting sperm then enter the lumen of the tubule. Thus, transcription from the male genome is not detected again until it is reactivated sometime during development.
So although there is some gene transcription in individual haploid spermatids, the fact that they are functionally diploid as a result of being connected by cytoplasmic bridges, and the fact that the long cell lineages result in many sperm all coming from the same spermatogonial stem cell, the answer to your question appears to be the latter......
Danniel said:
Or spermatogenesis is fully controlled by the genotype of the spermatozan's "owner", indifferently to the specific genetic content of each sperm cell, thus any sperm cell being part of its producer's phenotype?
......and that differences between individual sperm are the result of random <I>de novo</I> structural defects that occur in the "construction" of the motility apparatus during spermiogenesis.<P>
