Research Centers

photo

Schmidt Lab

Research topics include: Cyclin D1

Emmett V. Schmidt, MD, PhD
Professor of Pediatrics
Harvard Medical School

Vice Chairman for Education
Massachusetts General Hospital for Children

Research Summary

Cyclin D1 Causes Breast Cancer by Overexpression of its Normal mRNA and Protein

Cyclin D1 was initially identified at Massachusetts General Hospital as a translocationtarget in the parathyroid hormone locus in parathyroid tumors. Importantly, cyclin D1 was particularly interesting because it was the first cyclin linked by cytogenetic evidence to cancer. Given cytogenetic and expression studies showing that cyclin D1 was a key contributor on the 11q13 breast cancer amplicon, it became especially important to show that cyclin D1 could cause cancer. Our MMTV-cyclin D1 transgenic mice were the first clear demonstration that a cyclin could be an oncogene.

Genetic Interactions of Cyclin D1

Cyclin D1 is not required for cell division in knockout studies, indicating that it is primarily a regulatory molecule. It is the downstream connector to cell division in many signaling pathways. In particular, erbB2 signals through cyclin D1 in mouse mammary tumors. We examined this phenomenon in depth, using an MMTV-p16 travsgene to show that the p16INK4A tumpr suppressor that blocks cdk4/6 blocks tumor formation in MMTV-erbB2 mice. However, the situation is more complex in human tumors. We have shown a well-described dissociation between cyclin D1-expressing and erbB2-positive tumors, which is particularly noteworthy because cyclin D1 overexpression is narrowly confined to the ER positive subset of tumors. Importantly, our study suggested that erbB2 is more directly controlling the cell cycle through other mechanisms because its overexpression was non-redundant with cyclin D1 overexpression, loss of p16 or loss of pRb. To find translational correlates for our findings, we evaluated a therapeutic agent that blocks cdk4 activity, flavopiridol, and found that it was synergistic with trastuzumab in treatment of erbB2 positive breast cancers.

Since cyclin D1 likely functions in a particular genetic context, we have tested a variety of other possible genetic interactions. First, cyclin D1 works downstream of NFkB, especially in mammary development. Second, the oncoprotein kinase chaperone CDC37 functions as an oncogene in mice and collaborates with cyclin D1 in transformation of multiple tissues. We have found no evidence that it collaborates with p53 loss in tumorigenesis.

Growth and Cell Division - Regulation of Cyclin D1

The G1 cyclins are positioned in cell cycle control to monitor the cellular environment and determine the rate of passage through G1, after which cells become committed to completion of cell division. We have long been interested in connections between cell growth and cell division because cells must first grow before they can divide. We first became interested in this problem when we found that the c-myc oncogene works in part by controlling cell growth through regulation of the translation initiation factor eIF4E. To better evaluate the specificity of the effects of an otherwise general translation initiation factor we became interested in its ability to specifically enhance cyclin D1 protein levels without affecting cyclin D1 mRNA levels. This translational control has since been confirmed in many other labs, and appears to be related to effects on nuclear-cytoplasmic export of the cyclin D1 mRNA. We have further evaluated the expression patterns of cyclin D1 during pregnancy, and found an interesting cessation of cyclin D1 expression during mammary involution that may be important to the terminal differentiation of mammary glands after a first pregnancy.

Functions of Cyclin D1

Our study of cyclin D1 expression during mammary involution suggested that it may contribute to tumorigenesis by altering cell differentiation and demonstrated a significant function for p16INK4A in post-natal mammary differentiation. These regulatory mechanisms used during mammary involution offered a potential explanation for the protective effect of pregnancy against breast cancer. Our MMTV-cyclin D1 mice develop mammary tumors after a prolonged latency of nearly 18 months, but with an eventual penetrance of around 50%. The MMTV-transgenes effects on involution may be especially important in our model because no tumors have developed if a mouse has not gone through at least one pregnancy. Interestingly, our model is somewhat unique for mouse transgenics because the cyclin D1 tumors are estrogen receptor positive, like their human counterparts.

The long latency of MMTV-cyclin D1 induced tumor formation particularly implies that additional genetic events are required to collaborate in mammary carcinogenesis. We therefore used laser capture microdissection to attempt to define the genetic differences between non-invasive hyperplastic tissues driven by cyclin D1 expression and invasive tumor tissues in the MMTV-cyclin D1 mice. Importantly, we further tested the genes identified in mouse microarray studies in human breast cancers expressing cyclin D1 in LCM-isolated tissues from patient-matched normal, ductal carcinoma in situ, and invasive ductal carcinoma. We identified higher expression levels of immediate early response protein IEX-1, small stress protein 1 (HSPB8), and tumor necrosis factor-associated factor-interacting protein mRNAs in invasive lesions. These genes induced anchorage independence, increased cell proliferation, and protected against apoptosis, singly or in collaboration with erbB2. Importantly, they were all up-regulated by 17beta-estradiol and cyclin D1, and cyclin D1 overexpression increased p300/CBP binding to their promoters, supporting a novel model that cyclin D1-estrogen receptor (ER) coactivator interactions are important in ER-positive breast cancer. The heat shock protein B8 (HSPB8) is particularly associated with cyclin D1 and ER positive tumors, and we recently found that it can mediate cyclin D1s enhancement of radiation sensitivity. We are actively exploring the possible contributions of additional cyclin D1- and ER-regulated genes to its oncogenic functions.


Emmett V. Schmidt, MD, PhD
Principal Investigator

Group Members

  • Li Chen
  • Paul Yang
  • Cuiqi Li

Schmidt Laboratory

55 Fruit Street
Boston, MA 02129

Phone: 617-726-5707
Email: schmidt@helix.mgh.harvard.edu