Human Autoantibodies Reveal Titin as a Chromosomal Protein

A review by Alex Caldwell

    Because so few proteins have been found that are involved in chromosome structuring, their assembly has been relatively unknown.  Previous researchers have used autoantibodies to learn more about specific autoimmune diseases.  In this paper, the authors utilized human autoimmune sera of scleroderma to identify a protein involved in chromosome structure.  The serum from this human autoimmune disease binds to an epitope on mitotic chromosomes of human cells as well as Drosophila embryonic cells.  The authors theorized that the protein may be titin.
    Titin is a large protein whose I-band in vertebrates contains a region which is rich in proline, glutamic acid, valine, and lysine, the so-called PEVK domain.  The protein provides elasticity and structure for muscles in vertebrates and Drosophila.  The NH2 terminus of titin attaches to the COOH-terminal region of anti-actinin.  This binding effectively links titin to actin, another protein involved in muscle structure and contraction.  The authors wished to test whether titin was involved in chromosome condensation and structure and whether it is the same form of titin as in muscles performing this function.
    Because an epitope was recognized by the scleroderma serum, the authors screened a Drosophila expression library and located a gene which encodes for the Drosophila form of titin.  This was designated D-Titin.  The researchers proposed that this D-Titin was expressed in embryonic and striated muscles.
 Human epithelial cells (fixed HEp-2 cells) were stained.  They were incubated in primary antibody, washed with PBSTB, then incubated labeled fluorescently with secondary antibody before a second series of washings with PBSTB.  The cells were incubated in propidium iodide, which stained them red, and then washed with PBS.  In situ hybridization and staining was performed on the Drosophila embryos as described in the protocol.  Adult thoracic muscle and larval gut muscle were also immunostained and stained as described in the procedure from a previous paper. In Figure 1A, the human epithilial cells had almost identical staining with the scleroderma sera and the propidium iodide, which detects DNA, implying that the serum was indeed binding to the appropriate gene, which encodes the titin protein.  The Drosophila embryonic nuclei did not have the same staining pattern with the scleroderma and the propidium iodide.  The green staining with the scleroderma produced blurry spots in interphase, blobs in prophase, distinct shapes in metaphase (the same shape is produced with the propidium iodide), a different defined shape in anaphase (the same shape is produced with the propidium iodide), and a blur in telophase.  In contrast, the red propidium iodide generated spots in interphase, blurry spots in prophase, and distinct globs in telophase.  In metaphase and anaphase, the spots are very similar to the ones generated by the scleroderma sera and much yellow (signifying overlap) can be seen in the merged image.  There is also yellow in prophase and some in interphase.
    Figure 1B shows the chromosomal staining with an affinity purified polyclonal antibody, anti-LG.  This antibody was made using rabbits as the foreign host for the D-Titin protein.  The staining pattern of the HEp-2 was almost identical for the anti-LG probe and the propidium iodide.  In metaphase, the stained nuclear shape resembled one blob, while in anaphase the staining was of two nearby blobs.  For the Drosophila embryos, the staining was also the same between the anti-LG and propidium iodide.  In metaphase, it was again one blob but this blob had a more definite shape than the one for the Hep-2 cells.  The staining in anaphase showed two oval-like shapes, which were very close to each other.  The similarity between the banding patterns indicates that the anti-LG and the propidium iodide both bind to the same DNA.  The similarity between the Hep-2 cells and the Drosophila means that the DNA sequences are very similar.
    The anti-KZ antiserum was made in rats in a similar manner to that of the anti-LG serum in the rabbits.  The similar staining patterns of Figure 1C leads to the same conclusion as in Figure 1B.  The anti-KZ is binding to the same place as the propidium iodide proposing that the stains were marking the same sequences.
    Figure 2A is the restriction map of the genomic region of D-Titin.  The authors isolated the phage clones using flanking DNA to the areas of interest or were isolated directly using DNA from LG as a probe.  Figure 2B is a multiple sequence alignment for human titin, chicken titin, and Drosophila titin.  In Figure 2C, only the sequence that contains the PEVK-domain is given.  This is the sequence that was originally isolated with the scleroderma serum (which is also the LG clone).
    In order to test the expression of the gene that encodes the titin protein in Drosophila, embryos were stained dark blue by in situ hybridization and brown during immunostaining, shown in Figure 3.  Embryonic stages 10 and 11 (A/Aí) were the first to show any protein accumulation but did not exhibit detectable RNA.  In B/Bí, staining of the RNA first became visible and staining of the protein became more concentrated in the somatic mesoderm/musculature.  There was also a minute part in the pharynx.  As the embryos grew (C/Cí), the amount of staining also grew, with the dark blue covering almost the entire embryo.  The brown staining for the protein did not cover quite as much of the embryo but much more than during embryonic stage 13.  Panels D/Dí contained the stained embryos from embryonic stage 17.  RNA and protein are completely detectable in the fully developed musculature.  On the dorsal/ ventral side in E/Eí, RNA and protein is located in the visceral mesoderm/musculature and the somatic mesoderm/musculature.  In panels F/Fí, staining is evident in the visceral musculature and the somatic musculature on the dorsal/ventral sides.  At embryonic stage 15 (G/Gí), more musculature is evident, and the RNA staining is concentrated in the pharynx and somatic musculature.  The protein staining is located in approximately the same location as the RNA stain but also more towards the center of the embryo.  In embryonic stage 16 on the dorsal side, protein and RNA staining covers most of the embryo with coloring of the somatic and visceral musculature and less of the pharynx.  Figure 3 serves to show that D-Titin is exhibited in a developing Drosophila embryo, implying that it is present from the beginning, even for structuring the chromosomes.
    Figure 4A shows a phase image of a myofibril from an adult Drosophila thoracic muscle and then an image of the myofibril stained with the anti-KZ serum.  In panel B, the staining by the anti-KZ serum and the Texas red-phalloidin does not overlap, with the anti-KZ serum leaving a striped pattern.  Figure 4C shows some overlap between anti-KZ and scleroderma serum.  In panel D, most of the green ëstripesí overlap with the staining from the affinity-purified anti-LG.  Figures 4E and 4F show complete overlap with anti-KZ and human serum MIR and anti-KZ and anti-Zr5/Zr6, respectively.  Panel G is the staining pattern of the anti-KZ serum on the third instar larval gut muscle.  The short rectangular stained blocks were arranged almost in columns but scattered about the panel.  The overlap of E and F stains suggests the relationship between titin and actin and between the antibodies from the rabbit and the rat.
    Figure 5 is a western blot (except for lanes 1).  The protein extracts for (a) came from 8-24 h Drosophila embryos, 0-2 h Drosophila embryos for (b), and HeLa cells for (c).  Lanes 1 in (a)-(c) are simply blue-stained SDS-gels, which give protein size markers.  The other lanes are all from immunoblots which were cut into strips and then pieced together.  The following sera were added to the incubations: anti-LG to lane 2, LG preimmune serum to lane 3, anti-KZ to lane 4, and KZ preimmune serum to lane 5.  Lane 6 in (a) is a shorter exposure of the more developed embryos to the anti-KZ antiserum.  The most pronounced banding appeared in lanes 2a, 4a, and 6a.  There was also some banding in lanes 2 and 4 of (b) and (c).  The appearance of protein with anti-LG and anti-KZ sera is consistent with the data obtained in Figures 1B and 1C and 4E and 4F.
    The Hep-2 cells were stained with specific antibodies to vertebrate titin and with propidium iodide.  With N2A, the staining was similar for each stain but the propidium iodide was much more visible in the overlap.  The MIR antibody showed a high overlap towards the center of the condensed mitotic chromosome with more antibody present in a broader area.  Antibodies BD6, CE12, and A168 all had complete overlaps between the antibody and the propidium iodide.  These three antibodies bound to the titin in the A-band towards the COOH-terminus.  This figure indicates that the NH2 terminus of the titin differs between vertebrate titin and D-Titin.
    With these data, the authors concluded that titin does play a role in the structure of condensed chromosomes.  Previous experimenters had found that there seems to be a protein that effectively ëmeasuresí the length of the chromosome, which determines the chromosome condensation.  Machado et al. assert that titin performs this function in humans and Drosophila.  For future experiments, it might be interesting to see if a form of titin also fulfills this duty in other organisms, such as plants, other vertebrates, or invertebrates.  The same type of procedures would suffice to determine this function.  If titin does not operate in this capacity, then perhaps another structural protein specific to that group of organisms does.
    The existence of the PEVK domain raises questions about its function in the elasticity and structural components of titin.  If this region were removed (perhaps with homologous recombination) or altered, would this titin-mutant still function in a similar manner as the wild type titin?  Also, for organisms that do not have titin as their main structural protein, maybe their specific proteins have a region like that of titin which is rich in specific amino acids.  Is the repetitiveness vital to binding properties?  Why would TWITCHIN, another gene close to the titin family, not have an obvious PEVK domain?  What would other mutations of titin do to its function?  The same experiments could be executed and the data could be compared with the data obtained in this research.
    If D-Titin seemed to function in a comparable manner as vertebrate titin, then why would the vertebrate form have a longer sequence towards the 5í end?  Perhaps this titin performs another function that the D-Titin does not.  To answer this question, the authors could further pursue the question of why some antibodies did not bind and some sera did not stain the embryos and muscle tissues.
    Further experimentation could be carried out to determine the interaction between titin and other components of the condensed chromatin.  The interaction between titin and other proteins might also give information on their role in chromosome condensation.  Since Drosophila itself has at least three titin genes, maybe other organisms also have many genes in the titin family that perform other important functions in the chromosomes or muscles or other areas.  A difference in % gradient gels may also produce a slightly different banding pattern than the one seen in Figure 5.
    Since an autoimmune serum was used to identify titin, perhaps another autoimmune serum could identify a different protein in the same organisms or in entirely new groups of organisms.  The procedure for this experiment would be similar to the one followed by these researchers.  Perhaps the identification of other proteins with autoimmune sera could help diminish the number of autoimmune diseases.
    The authors are able to prove that titin does have a role in chromosome condensation through the figures and the relationship between titin and several autoimmune sera.  Further roles of titin need to be investigated, as well as the specific regions at which titin is centered on the chromosomes.  However, the size of the pertinent protein on the chromosomes, the distinctiveness of the sequence, and the articulation of the protein in developing embryos and human cells all lead to the conclusion that titin is the protein in question and its part in muscle elasticity and chromosome condensation is apparent and distinct.

Return to Main Page

Return to Molecular Page

e-mail me: