Cell migration requires coordinated reorganization of cytoskeleton, lamellipodial extension, formation of forward adhesions, exertion of contractile force to pull the cell body forward, and the detachment of the rear 
. α-Actinins originally discovered as actin linking proteins have been shown to play a key role in regulating cell migration. In the past decades, accumulating evidences revealed that α-actinins interact with a large number of proteins including many transcription factors, in addition to actin 
. While both α-actinin-1 and α-actinin-4, two non-muscle isoforms have been shown to localize along the stress fiber, α-actinin-4 also accumulates at the leading edge of invading cancer cells 
and motile fibroblasts 
. This is consistent with α-actinin-4 being reported to play a crucial role in cancer invasion and metastasis 
In this study, we found a counter-intuitive cellular function of α-actinin-4 in fibroblasts while dissecting its role in migration; we used a stable cell line of murine lung fibroblasts in which α-actinin-4 has been almost completely eliminated. In line with previous reports 
, we found that knockdown of α-actinin-4 significantly impaired the cell motility of murine lung fibroblasts. Furthermore, our results showed that the impaired migration of fibroblasts was probably due to the scarcity of focal adhesion within cell body of ACTN4 KD fibroblasts and the elongation of focal adhesion sites at the periphery. The impaired focal adhesion of ACTN4 KD fibroblasts was not due to the alteration in level of the adhesion molecule vinculin (). However, we could not directly attribute the migration to the number or location of adhesions. As adhesion assembly and turnover are highly dynamic and orchestrated processes essential for cell migration 
, we determined if the impaired cell migration of ACTN4 KD fibroblasts was due to the decreased turnover rate of focal adhesion at periphery. We found that the knockdown of ACTN4 did not affect turnover of focal adhesion (data not shown). However, the impaired cell migration of ACTN4 KD fibroblasts may also be due to the increased number of protrusions in ACTN4 KD fibroblasts as cells were impaired in their ability to establish a cell polarity for productive locomotion.
Recently, Choi and his colleagues have found that Myosin II is required for the elongation and maturation of adhesion 
. Thus we sought to determine if the enlarged focal adhesion sites at the periphery of ACTN4 KD fibroblasts were caused by the increase of myosin II expression. As shown in , our qPCR result showed that knockdown of ACTN4 did not affect the expression of myosin II in fibroblasts although the amount of myosin II mRNA was low. To our surprise, MYH9 and MYH10 in WT ACTN4 and ACTN4 KD fibroblasts are abundant and both MYH9 and MYH10 co-immunoprecipitate with an ACTN4 mutant (unpublished data). Knockdown of ACTN4 significantly increased the expression of MYH10 but not MYH9 at protein level instead of mRNA transcription (). MYH9 and MYH10 probably substitute for myosin IIA and IIB, respectively. The increase of MYH10 expression in ACTN4 KD fibroblasts may cause the formation of enlarged focal adhesion sites at the periphery although the mechanism remains to be determined.
Recent reports suggested that the ACTN4 may play an important role in transcription regulation 
. Indeed, current findings 
and our results showed that the downregulation of ACTN4 significantly affected the expression of myosin II in Glioma and MYH10 and myosin light chain (MLC2) in fibroblasts. The increase of traction strength of the KD cells was likely due to either reduced actin filament bundling 
or enhanced expression of MLC2, a contractile machinery element. The adhesive strength at one hour after plating was less in these cells (), but that is consistent with the increased traction strength as this would counter the ability to establish a large number of adherent sites shortly after attachment.
Our results showed the opposite cellular function of α-actinin-4 on focal adhesion and contractile force in fibroblasts compared to those reported in glioma cells suggesting that the integrated cellular functional consequences of α-actinin-4 activities are partially dictated by the cell type. Although the downregulation of ACTN4 caused a significantly increase in traction force of ACTN4 KD fibroblasts, these cells migrated slowly relative to WT ACTN4 fibroblasts. This was probably due to two reasons. First, the direction of all traction force faced towards the central cell body from the cellular periphery instead of facing one direction. Second, ACTN4 KD fibroblasts produced several “tail-like” structures but failed to produce a dominant protrusion (or cell polarity) although its focal adhesions within the cell body were impaired. In order to determine if the impaired migration of KD cells was partially due to the reduced proliferation (), we measured the motility of KD cells quiesced with quiescence medium in the absence of FBS for 24 hr by performing a standard scraped wound healing assay and found that quiesced KD cells still migrated significantly slower than WT cells (data not shown). This suggested that the impaired motility of KD cells was not caused by the reduced proliferation.
Our findings also suggested that the ACTN4 is essential for maintaining normal cellular and nuclear cross-sectional area. However, the mechanisms of how ACTN4 plays an important role in controlling cellular and nuclear cross-sectional area are still unclear. For the first time, our results also showed that the knockdown of α-actinin-4 also significantly impaired fibroblasts proliferation. This is not surprising as several reports in the literature revealed a function for α-actinin-4 in cytokinesis during mitosis 
One question remains as to why there are multiple α-actinin isoforms. Although the identify of amino acids between α-actinin-1 and α-actinin-4 is 87%, these two isoforms are reported to play contrasting roles in cancer cell survival and the effect of ACTN4 shRNA on cell growth even depends on the cell type 
. Still, in our hands, we could restore the normal cell phenotype by expressing ACTN1 in lieu of ACTN4, suggesting that these two proteins are potentially interchangeable. This is a conundrum in that in the absence of ACTN4, the endogenous ACTN1 did not prevent the phenotypic changes, nor was ACTN1 compensatorilly upregulated. Thus, either there is a protein dosage aspect that was restored by expression of either ACTN4 or ACTN1, or the forced expression of ACTN1 led to aberrant cellular functioning. As a further examination of this potential redundancy, we attempted, but failed to sufficiently knockdown Actinin-1 in KD fibroblasts by siRNA; this was probably due to the long half life of actinin (more than 30 hrs) and the low transfection efficiency of primary fibroblasts. Thus, this question of isoform functionality remains for future investigations.
In conclusion, our findings show that the α-actinin-4 is essential for maintaining the normal cell morphology, cellular and nuclear cross-sectional area, focal adhesion, contractile force and hence motility. The appropriate expression of α-actinin-4 is also important for normal proliferation of fibroblasts. Future studies should investigate why knockdown of α-actinin-4 affects the expression of myosin II and myosin light chain 2 in cancer cell and fibroblasts.